UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The role of connexin43 in glioma motility Bates, David Christopher 2006

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

Item Metadata


831-ubc_2006-0370.pdf [ 7.11MB ]
JSON: 831-1.0092669.json
JSON-LD: 831-1.0092669-ld.json
RDF/XML (Pretty): 831-1.0092669-rdf.xml
RDF/JSON: 831-1.0092669-rdf.json
Turtle: 831-1.0092669-turtle.txt
N-Triples: 831-1.0092669-rdf-ntriples.txt
Original Record: 831-1.0092669-source.json
Full Text

Full Text

THE R O L E OF CONNEXIN43 IN G L I O M A M O T I L I T Y  By DAVID CHRISTOPHER BATES B . S c , The University o f Western Ontario, 2004  A THESIS S U B M I T T E D I N P A R T I A L F U L F I L L M E N T OF THE REQUIREMENTS FOR THE D E G R E E OF  M A S T E R OF SCIENCE in  THE F A C U L T Y OF G R A D U A T E STUDIES  (Anatomy and Cell Biology)  THE UNIVERSITY OF BRITISH C O L U M B I A September 2006 © David Christopher Bates 2006  ABSTRACT  Gap junctions, proteinaceous channels which directly link the cytosol o f adjacent cells and allow the passage o f ions and small molecules, are formed from the hexameric oligomerization o f connexin subunits. W e are interested in the role o f Connexin43 (Cx43), the most abundant isoform expressed in astrocytes, in glioma motility. T o achieve this objective, we have isolated a C 6 subclone endogenously expressing high levels o f Cx43 (C6-H) and have employed in vitro wound healing and transwell assays to evaluate cellular motility. When compared to parental C 6 cells in which Cx43 is expressed at low levels (C6-L), the C 6 - H subclones were more motile. To deduce whether Cx43 was indeed responsible for the observed differences in motility, the C 6 - H cells were retrovirally infected with Cx43 s h R N A to stably knock down Cx43 expression. Coincident with the knockdown o f endogenous C x 4 3 , a decrease in motility and invasion was observed. A s gap junction intercellular communication (GJIC) was also decreased, motility assays were conducted in the presence o f gap junction inhibitors to evaluate the contribution o f G J I C to cell motility. Because no significant differences in motility could be detected upon blocking G J I C , C 6 cells exogenously expressing full length or truncated Cx43 were subjected to the aforementioned motility assays to expose alternate mechanisms o f Cx43-mediated motility. Cells expressing full length Cx43 exhibited increased motility while cells expressing the truncated form of Cx43 did not. Our results indicate that downregulation o f Cx43 decreases motility in C 6 glioma cells and suggest that the carboxy terminus plays an important role in Cx43-mediated motility. Keywords: Gap junction, Connexin43, C 6 cells, Glioma, Motility, Invasion, R N A i .  ii  TABLE OF CONTENTS Page Title Abstract Table o f contents List o f Tables List o f Figures List o f Abbreviations Dedication Acknowledgements  i ii iii v vi viii xi xii  CHAPTER 1 INTRODUCTION 1.0 Motility 1.1 Directional vs. Random motility 1.2.1 A c t i n Polymerization 1.2.2 A c t i n Stabilization 1.3 Gap Junctions 1.4 Connexins 1.5 Connexin43 1.5.1 Cx43 Life cycle Regulation o f Cx43 Cx43 Trafficking and Plaque Formation 1.6 Cx43 Associating Proteins 1.6.1 Overview 1.6.2 Cadherins N-Cadherin 1.6.3 Catenins p l 2 0 catenin 1.6.4 Drebrin 1.6.5 Nephroblastoma Overexpressed 1.7 Involvement o f Cx43 in C e l l Motility 1.7.1 Mechanisms o f Cx43-Mediated Motility 1.8 Cx43 in Cancer 1.9 Brain Cancer 1.9.1 Glioma 1.9.2 Nature o f Gliomas 1.9.3 Causes 1.9.4 Treatment/management strategies 1.10 C 6 cells 1.11 Rationale  1 1 2 3 3 4 6 7 7 8 9 9 ....9 10 11 11 11 12 12 13 14 15 15 16 17 18 19 20  CHAPTER 2 MATERIALS A N D METHODS 2.1 C e l l Culture 2.2 Gap junction blocker treatments 2.3 Immunocytochemistry 2.4 Fluorescence Microscopy (Image acquisition) 2.5 Protein Collection iii  28 28 28 29 29  Table of Contents cont'd  Page  2.6 S D S - P A G E 2.7 Immunoblotting 2.8 Preloading/Dye Coupling : 2.9 W o u n d healing motility assay 2.10 Transfection and infection of s h R N A 2.11 Transwell migration invasion assays 2.12 Statistical analysis CHAPTER 3 RESULTS 3.1 Motility o f C 6 - H subclones 3.2 Knockdown of Cx43 in C 6 - H subclones by s h R N A 3.3 Knockdown of Cx43 by s h R N A attenuates cell motility and invasion 3.4 Mechanisms of Cx43 Mediated motility 3.4.1 Protein localization and gap junctional intercellular Communication in C 6 - C x 4 3 A C T 2 4 4 - 3 8 2 G F P cells 3.4.2 Expression of full length Cx43 enhances directional motility 3.4.3 Cx43 associating proteins i n cells expressing Cx43A244-382GFP  30 30 31 31 32 32 33  34 34 35 36 37 38 38  CHAPTER 4 DISCUSSION 4.1 Summary 4.2 Cx43-Mediated Invasion in Homocellular Populations o f C 6 Cells 4.3 Cx43 and G J I C in G l i o m a Chemokinesis and Chemotaxis 4.4 Involvement of G J I C 4.5 Involvement of the Cx43 C-Terminus in Chemokinesis 4.6 Involvement of the Cx43 C-Terminus in Chemotaxis 4.7 The Role of Cx43 in C e l l Motility C o u l d B e Cell-Type Specific 4.8 Conclusions and Physiological Relevance 4.9 Future Directions  82 84 86 87 89 91 92 92 94  REFERENCES  95  iv  LIST O F T A B L E S  Page Table 1.  Implication of Gap Junctions in Disease  v  25  LIST OF FIGURES Page Figure 1.1  A c t i n polymerization  22  Figure 1.2  Connexins, Connexons, and Gap Junction Formation  24  Figure 1.3  Types o f Gap Junction Channels  27  Figure 3.4  Connexin43 expression in C 6 subclones  Figure 3.5  Motility and Invasion assays  43  Figure 3.6  Analysis o f C 6 subclone motility by wound healing  45  Figure 3.7  Analysis of C 6 subclone motility by transwell  47  Figure 3.8  Efficiency of endogenous Cx43 knockdown in C 6 cells  49  Figure 3.9  Knockdown of endogenous Cx43 in C 6 cells  51  Figure 3.10  Knockdown of Cx43 by s h R N A decreases directional motility  53  Figure 3.11  Knockdown of Cx43 by s h R N A decreases non-directional motility  55  Figure 3.12  Knockdown of Cx43 by s h R N A decreases invasion  57  Figure 3.13  Knockdown of Cx43 by s h R N A reduces gap junction intercellular  40,41  Communication  59  Figure 3.14  Efficacy of the gap junction blocker carbenoxolone  61  Figure 3.15  B l o c k i n g gap junction intercellular communication did not alter directional motility  Figure 3.16  63  B l o c k i n g gap junction intercellular communication did not alter nondirectional motility  Figure 3.17  65  Levels o f exogenous Cx43 are similar between cells expressing truncated and full length forms of Cx43  67  Figure 3.18  Localization o f truncated Cx43 in C 6 cells  69  Figure 3.19  C 6 cells expressing truncated Cx43 are communication competent  71  vi  List c'  cont'd  Page  Figure 3.20  Enhanced directional motility is conferred by the carboxy terminus ofCx43  Figure 3.21  73  The carboxy terminus of Cx43 does not enhance non-directional Motility  Figure 3.22  75  Deletion of the Cx43 carboxy terminus results in a more adhesive Phenotype  Figure 3.23  77  Expression and localization o f N-Cadherin is not altered in C 6 cells expressing C x 4 3 A C T - G F P  Figure 3.24  79  Expression and localization o f pl20ctn is not altered in C 6 cells expressing C x 4 3 A C T - G F P  81  vii  LIST OF ABBREVIATIONS AM  Acetoxymethyl  ANOVA  Analysis o f Variance  APC  Adenomatous Polyposis C o l i  ARP2/3  Actin-Related Protein 2/3  BMP(s)  Bone Morphogenic Protein(s)  BRMS1  Breast Cancer Metastasis Suppressor 1  C-  Carboxy  DBT2  Delayed Brain Tumor  DIC  Differential Interference Contrast  DMEM  Dulbecco's M i n i m a l Essential M e d i u m  CBX  Carbenoxolone  CCN  Cysteine-Rich61 /Connective Tissue Growth Factor/Nephroblastoma  Cdc42  C e l l Division Cycle 42  CK1  Casein Kinase 1  CMV  Cytomegalovirus  CMV43  M i c e in which Cx43 is overexpressed by the C M V promoter  CNS  Central Nervous System  CNTF  Ciliary Neurotrophic Factor  CNTFsRa  C N T F soluble receptor alpha  Cx(s)  Connexin(s)  Cx43  Connexin43  Cx43KO  Cx43 knockout  DADS  D i a l l y l Disulfide  DBMA  Dimethylbenzanthracene viii  wiations cont'd DIC  Differential Interference Contrast  EGFR  Epidermal Growth Factor Receptor  EMT  Endothelial-to-Mesenchymal Transition  ER  Endoplasmic Reticulum  ERK  Extracellular Signal Related Protein Kinase  F-  Filamentous  G-  Globular  GAPDH  Glyseraldehyde-3-phosphate dehydrogenase  GBM  Glioblastoma Multiforme  GCV  Ganciclovir  GFP  Green Fluorescence Protein  GJIC  Gap Junction Intercellular Communication  GZA  Glycyrrhizic A c i d  HRP  Horse Radish Peroxidase  HSPC300  Heat Shock Protein C300  HSVtk  Herpes Simplex Thymidine Kinase  LIMK  L I M kinase  MAPK  Mitogen Activated Protein Kinase  MMP  Matrix Metalloprotease  mRNA  Messenger R N A  N-  Amino  Napl25  Nck-associated protein 125  N-Cad  Neural Cadherin, N-Cadherin  NC  Neural Crest ix  NF1/2  Neurofibromatosis  1/2  NOV  Nephroblastoma  PAGE  Polyacrylamide G e l Electrophoresis  pl20ctn  p i 2 0 catenin  PET  Polyethylene Terephthalate  PDGF  Platelet-Derived Growth Factor P D G F R  PDGFR  Platelet-Derived Growth Factor Receptor P D G F R  PKA  Protein Kinase A  PKC  Protein Kinase C  PTC  Papillary Thyroid Carcinoma  RIPA  Radioimmune Precipitation Lysis Buffer  RNA  Ribonucleic A c i d  SH2  Src-Homology 2  shRNA  Short Hairpin R N A  siRNA  Short Interfering R N A  TBS  Tris-Buffered Saline  TBST  Tris-Buffered Saline containing Tween 20  TP53  Tumor Protein53  TSC2  Tuberous Sclerosis Complex-2  VASP  Vasodilator-Stimulated Phosphoprotein  WASP  Wiskott-Aldrich Syndrome Protein  WAVE  W A S P Family Verprolin Homologous Protein  WHO  W o r l d Health Organization  ZO-1  Zona Occludens-1  Overexpressed  X  DEDICATION  1  THIS WORK IS DEDICATED TO M Y G R A N D A D , T H O M A S B A T E S , AND TO HIS WORDS OF WISDOM: 'BE GREEDY TO L E A R N ' AND 'EVERYTHING BECOMES A M E M O R Y '  xi  ACKNOWLEDGEMENTS I would like to express sincere gratitude to Dr. Christian Naus, particularly for his guidance and financial support throughout m y studies and additionally for providing me with the opportunity to present my work internationally.  I am indebted to Dr. W u n Chey Sin and John Bechberger for their ongoing advice and technical assistance. Props extend to all other members o f the Naus lab - M i k e Kozoriz, Charles L a i , Steve Bond, Lynne Bechberger, numerous work study students and volunteers, and even C i m a Cina.  Members o f m y advisory committee Dr. Robert N a b i and Dr. J i m Johnson provided me with much appreciated guidance and have challenged me such that I could not but become a scientist.  This work would not have been possible without the generosity o f Dr. Dale Laird and Dr. Qing Shao who provided several constructs employed throughout these studies.  I must acknowledge Dr. Greg K e l l y without whose inspiration m y scientific career would never have begun.  A s it takes a village to raise a child, m y family and friends who are m y village can not go unrecognized for their involvement in this work.  The love, patience, and support o f m y Jing throughout these studies have always meant more to me than any o f the work, and it is with all o f m y heart that I offer her m y appreciation.  Finally, I am grateful to anyone who reads the following pages -consider them a journal of m y life for the past two years. Godspeed.  xii  1  CHAPTER 1 INTRODUCTION 1.0 Motility C e l l motility is an essential process for unicellular and multicellular organisms alike as it underlies processes fundamental for survival. In addition to its role in development and physiological processes, such as inflammation and wound repair, cell motility plays a key role in pathophysiology o f tumorigenesis.  1.1 Directional vs. Random motility The distinction between chemotaxis or chemorepulsion, directional movement along or against a concentration gradient o f signal molecules, and chemokinesis, non-directional movement triggered by uniformly distributed signal molecules, enables the deconvolution o f the highly complex biological process we generalize as cell motility (Maheshwari and Lauffenburger, 1998). Note that the terms chemotaxis and chemorepulsion, although not equal, are both forms o f directional motility and w i l l be used interchangeably throughout this text. When directional or non-directional motility involves extracellular matrix ( E C M ) proteins or cell membranes, the terms haptotaxis and haptokinesis, respectively, are employed, see Anand-Apte and Zetter, 1997). The fundamental difference between directional and random motility that indeed allows their distinction is the acquisition and persistence o f cell polarity. Directional motility, which is persistence o f locomotion in a particular orientation, requires that the concentration o f a signaling molecule be sufficiently different along the body o f a cell that such differences are distinguishable and thereby confer orientation to the cell, and additionally requires that the cell is able to recruit the appropriate cellular machinery to respond to these differences (Anand-Apte and Zetter, 1997). B y contrast, cell polarization i n random motility is only transient and is governed by intrinsic factors (Wilkinson, 1998). Motility requires contraction and ultimately, cells must form adhesions with a substratum i n order for a contractile process to occur. Therefore it is the cell: substratum contact area, the number and strength o f  2 adhesions both to the substratum as well as to the cytoskeleton, the structure o f the cytoskeleton, and the force generated by the contractile machinery which enable contraction (Maheshwari and Lauffenburger, 1998). This implies that, front vs. rear asymmetry o f adhesions and their properties (i.e. expression, function, affinity), regardless o f how such asymmetry is established, is essential for a resultant net difference in traction and hence movement rather than isometric contraction (Huttenlocher et al., 1995; Lauffenburger and Horwitz, 1996). 1.2.1 A c t i n P o l y m e r i z a t i o n The process o f actin polymerization, the current paradigm for cell migration, is complex and involves multiple biochemical pathways (Figure 1.1). Mechanisms o f actin polymerization involve Rac, cell division cycle 42 (cdc42), and R h o A , which are all subtypes o f Rho GTPases within the Ras superfamily. The activation o f both Rac and cdc42 are elicited by a multitude o f signaling cascades and effector proteins. Activated Rac and the adaptor protein N e k cause the dissociation o f S c a r / W A V E ( W A S P (Wiskott-Aldrich Syndrome Protein )-family verprolin homologous protein) from a complex composed o f heat shock protein C300 (HSPC300), A b i , Sra-1, and Nck-associated protein 125 (Napl25) (Eden et al., 2002) while activated cdc42 activates W A S P and N - W A S P (Vicente-Manzanares et al., 2005). W A V E / S c a r - H S P C 3 0 0 and W A S P / N - W A S P both activate the actin-related protein 2/3 (ARP2/3) complex which binds to the pointed end o f F-actin and causes extension o f nascent filamentous (F)-actin at a 70° angle (Mullins et al., 1998). Activated cdc42 also activates the formin mDia2 which, when associated with monomeric, globular (G)-actin bound profilin, binds to the pointed end o f F-actin and causes extension o f F-actin in a linear fashion (Vicente-Manzanares et al., 2005). R h o A , is similarly governed by a variety o f signaling cascades and effector proteins, and results in the activation o f the formin m D i a l which, when associated with G-actin bound profilin, also binds to the pointed end o f F-actin and causes extension o f F-actin in a linear fashion (Kobielak et al.,  3 2004; L i and Higgs, 2003). The interaction o f vasodilator-stimulated phosphoprotein ( V A S P ) with vinculin is also important for actin-filament assembly (Brindle et al., 1996). 1.2.2 A c t i n Stabilization The stability o f F-actin is dependent not only upon polymerization, but also on disassembly. Both Rac and R h o A activate L I M kinase ( L I M K ) which in turn phosphorylates and inactivates cofilin leading to the disassembly o f F-actin into G-actin. Ena/ V A S P associated with profilin interact with Z y m i n and a-actinin to stabilize the barbed end o f F-actin as does the capping protein gelsolin, which is under the control o f phosphoinositides (Janmey and Stossel, 1987; K r a u s e e t a l . , 2003). 1.3 G a p Junctions The ability o f cells to communicate is o f fundamental importance to their survival. Not surprisingly, many different forms o f cellular communication exist. One form o f cellular communication is mediated by gap junctions, membrane-spanning channels which are formed by the association o f hemi channels (connexons) from adjacent cells (Figure 1.2). This association connects the cytosol o f adjacent cells allowing the selective and regulated movement/exchange of amino acids, ions, small molecules and metabolites (<lkDa) between cells (Simon and Goodenough, 1998; Spray and Bennett, 1985). Recently it has been shown that s i R N A is also able to pass between neighbouring cells v i a gap junctions, suggesting that gap junctions may play a role in the transfer o f other forms o f R N A such as m i c r o R N A (Valiunas et al., 2005). Cells which communicate in this manner are said to be coupled. Gap junction intercellular communication (GJIC) is essential for the viability o f numerous tissues and the selective and regulated movement o f gap junction permeable molecules is essential for proper vertebrate and invertebrate development and homeostasis (Reaume et al., 1995; L o and Gilula, 1979; Ackert et al., 2001; Mercola and Levin, 2001).  4  Gap junctions also electrically couple adjoining cells (Bennett, 1997). Electrical communication v i a gap junctions is especially relevant to cardiac and neuronal activity. Cardiac gap junctions mediate the flow o f current from the nodal pacemaker cells to ventricular myocytes which permits synchronous, rhythmic contraction (Herve and Sarrouilhe, 2006). In the brain, the synchronous, rhythmic spread o f action potentials throughout neuronal syncytia, likely initiated by the thalamus, is defined and realized by gap junction expression and activity respectively. Normal synchronous activity is important for physiological processes such as memory and sleep (Destexhe and Sejnowski, 2003) while pathological synchrony is the hallmark o f seizures (Fenner and Hass, 1989; Perez Velazquez and Carlen, 2000). A s gap junctions play a major role in vertebrate development, homeostasis, and electrical synchronizations, it is not surprising that perturbations in their expression and/or function have been implicated in many diseases (See Table 1). Such diseases include but are not limited to epilepsy/epileptogenesis (Perez Velazquez and Carlen, 2000), Charcot Marie Tooth Disease (Bergoffen et al., 1993) non-syndromic sensorineural deafness (Kelsell et al., 1997), cataracts (Shiels et al., 1998), Keratitis-ichthyosis-deafness syndrome (Richard et al., 2002), oculodentodigital dysplasia (Paznekas et al., 2003) and a variety o f cancers including ovarian, liver, breast, brain (Fentiman et al., 1979; Hoffman et al., 1993; Loewenstein and Kanno, 1966; Zhu et al., 1991); (See Mesnil et al., 2005) for a review o f gap junctions in cancer). 1.4 Connexins The identification o f gap junction constituent proteins, connexins (Cxs; Beyer et al., 1990), has offered much insight into the regulation and function o f gap junctions. In addition to numerous species specific isoforms, 21 connexin (Cx) family members have been identified in humans thus far (Willecke et al., 2002). T w o different nomenclatures exist to categorize these isoforms. The most common nomenclature uses the apparent molecular weight o f the proteins on sodium dodecyl sulphate ( S D S ) - P A G E (polyacrylamide gel electrophoresis) mobility assays to  5 identify individual connexins (Beyer et al., 1990). This nomenclature w i l l be used throughout the current study. Alternatively, individual connexins are subdivided based on structural similarity and differences into 3 types: a, p and y or 8 (Manthey et al., 1999). A l l members o f the connexin family have 4 highly conserved alpha-helical hydrophobic domains which span the membrane and line the pore o f connexons and gap junctions (Goodenough et al., 1996; Kelsell et al., 2001). Two extracellular loop domains mediate docking o f adjacent connexins (when oligomerized into connexons) resulting in the formation o f gap junction channels. In addition to a single cytoplasmic loop, both the amino (N) and carboxy (C) termini are cytoplasmic-facing. These cytoplasmic components are the moieties sensitive to p H and phosphorylation, important determinants for channel properties and function (Bruzzone et al., 1996). The C-terminus exhibits the highest variation in amino acid sequence and is therefore thought to be responsible for unique channel properties and function conferred by individual connexins. The noncovalent oligomerization o f connexins into hexameric membrane spanning connexons can occur in several ways (see Figure 1.3) with a vast number o f possible combinations. Identical connexins can oligomerize with one another to form homomeric connexons. The docking o f identical homomeric connexons form homotypic junctions while non-identical associations between homomeric connexons form heterotypic junctions. N o n identical connexins can also oligomerize with one another to form heteromeric connexons. When a heteromeric connexon associates with a homomeric connexon, a heteromeric heterotypic junction is formed while the association o f two heteromeric connexons forms a heteromeric homotypic junction ( C x - C x interactions are reviewed in (Herve et al., 2004a). The various combinations o f connexon associations and the compliment o f connexins that form an individual connexon alter channel properties such as selectivity and conductivity (White et al., 1995) and hence individual members o f the C x family can be studied independently o f one another or in various combinations (Brink et al., 1997). The physiological consequences o f differing channel  6 properties are particularly relevant since a given cell type can express a distinct yet variable compliment o f connexins, and therefore both compartmentalization and syncytia can be realized. For example, granulosa cells express both Cx43 and C x 3 7 . Direct homocellular interactions between granulosa cells are mediated by Cx43 (Gittens et al., 2003; Ackert et al., 2001) while heterocellular interactions at the granulosa-oocyte interface are mediated by Cx37 (Veitch et al., 2004; Simon et al., 1997). The differential permeability properties o f Cx43 and Cx37 suggest that the granulosa syncytium may be a compartment distinct from that o f the oocyte and thus enable discrete and independent signaling within the developing follicle (Elfgang et al.,1995; Gittens and Kidder, 2005). 1.5 Connexin43 Connexin43 (Cx43) is the most prevalent and well studied o f the isoforms within the C x multigene family. In cancer research, Cx43 is best known for its growth suppressive effect: it is downregulated in many types o f tumor tissue (see section 1.8) and its reconstitution leads to a reduced proliferation rate and tumor formation (Naus et al., 1992; K i n g et al., 2002). These findings were previously explained by G J I C alone however, more recently channel independent effects have been observed (Omori and Yamasaki, 1998; Huang et al., 1998c; Zhang et al., 2003b; L i n et al., 2003b). Indeed cells transfected with Cx43 have a different gene expression profile than mock transfected cells that may account for tumor suppression partially or wholly (Naus et al., 2000). The variable C-terminus o f Cx43 is perhaps the best candidate to explain non-junctional effects. A t 151aa long (40% o f the full length protein) it is the longest region o f variability among connexin isoforms. The C-terminus has 14 identified sites o f phosphorylation regulated by 6 kinase proteins (v-src, M A P K , PKC,p34cdc2, C K 1 , PKA-dependent; for reviews see Lampe and L a u , 2000; Lampe and Lau, 2004). The phosphorylation profile o f Cx43 is implicated in junctional conductance and channel gating, protein turnover, and modulates  7  interaction with many of the associating proteins identified thus far (Herve et al., 2004b; Giepmans, 2004). 1.5.1 C x 4 3 L i f e cycle Regulation of C x 4 3  Connexins can be regulated by endogenous and exogenous factors at several levels including transcription, translation, and posttranslational modifications. The promoter region of the Gjal gene which encodes Cx43 is regulated by numerous transcription factors (i.e. c-fos, cjun, C R E B , Spl) which bind to AP-1, AP-2, Spl, p53, and C R E sites (Sullivan et al., 1993; Y u et al., 1994) as well as other positive and negative regulatory sites (Chen et al., 1995). Activation of such transcription factors is realized by several identified factors including: phorbol esters, protein kinase C (PKC; Geimonen et al., 1996) ciliary neurotrophic factor (CNTF) in combination with it's soluble receptor CNTFsRa (Ozog et al., 2002; Ozog et al., 2004), bone morphogenic proteins (BMPs) (BMP2 and BMP4; Chatterjee et al., 2003; Bani-Yaghoub et al., 2000), W n t l , likely via T-cell factor (TCF)/LEF (lymphoid enhancer factor) binding elements (Ai et al., 2000; van der Heyden et al., 1998), and retinoids (through the RXR-P receptors (Batias et al., 2000). Translation of Cx43 mRNA is facilitated by a strong internal ribosomal entry site (IRES; Schiavi et al., 1999). The posttranslational modification of Cx43 is regulated by numerous biochemical signaling pathways, including kinases and phosphatases, but importantly the life cycle of Cx43 is a major site for the regulation of Cx43 expression. The half life for Cx43 is 1.35h (Beardslee et al., 1998; Berthoud et al., 2004; Darrow et al., 1995; Laird et al., 1991) and therefore the expression and persistence of gap junction plaques represents a balance between stabilizing mechanisms such as phosphorylation and degradation pathways. Indeed the upregulation of Cx43 by inhibitors of both protein synthesis and the proteasome underscores the important physiological role for Cx43 in maintaining cellular homeostasis, particularly when the  8  viability o f the cell is compromised (Musil et al., 2000). Furthermore, the disappearance o f Cx26 m R N A is completely inhibited by the protein synthesis inhibitor cycloheximide (Kren et al., 1993). It is possible that in addition to Cx43 protein, Cx43 m R N A might also be stabilized by the inhibition o f protein synthesis, supporting Cx43 stabilization when protein synthesis is compromised. C x 4 3 T r a f f i c k i n g and Plaque formation Cx43 is cotranslationally inserted into the endoplasmic reticulum ( E R ) membrane (White et al., 1995) and oligomerized into hexameric connexons in the trans G o l g i compartment (Bennett and Zukin, 2004; Das Sarma et al., 2001; M u s i l and Goodenough, 1993; Saez et al., 2003). However, the exogenous overexpression o f Cx43 (VanSlyke, 2005), induces premature assembly o f connexons in the E R . The aggregation o f connexons (10,000/um ; L a u f et al., 2002) z  in the plasma membrane forms gap junction plaques (Bruzzone et al., 1996) which are important for the opening o f channels and functional G J I C (Bukauskas et al., 2000). Connexons are delivered via vesicular transport along microtubules to the plasma membrane (Lauf et al., 2002). Note that the formation o f a Cx43/zona occludens-1 (ZO-l)/(3-catenin complex is required to target Cx43 to the plasma membrane in rat cardiomyocytes ( W u et al., 2003). Although mictotubules directly interact with Cx43 at gap junction plaques (Giepmans et al., 2001), connexons are transported to nonjunctional plasma membrane rather than to the center o f gap junction plaques (Lauf et al., 2002). One proposed mechanism for the plasma membrane localization o f individual connexons to gap junction plaques involves Gj proteins as their a  inhibition by pertussis toxin reduced plasma membrane levels o f Cx43 specifically in lowdensity lipoprotein (Lampe et al., 2001).  9 1.6 Cx43 Associating Proteins 1.6.1 Overview Recently, Cxs have received attention as proteins in their own right, independent o f gap junction channel formation. In addition to their interaction with other C x family members via extracellular loop domains, Cx43 associates with a variety o f other proteins, the majority o f which associate with the cytosolic C-terminus o f Cx43. The variable homology domain o f Cx43 contains 14 phosphorylation sites which are important for S H 2 interactions (Lampe and L a u , 2004). Cx43 associating proteins identified thus far include: Z O - 1 , Occludin, L i n - 7 , P-catenin, a-catenin, pl20ctn, pp60c-src, pp60v-src, P K C e , P K C a , D M P K , M A P K , P 3 8 M A P K , N - C a d , N O V , F-actin, actin, drebrin, clathrin, Caveolin-1, ubiquitin, a-tubulin, P-tubulin, CIP62, and numerous connexins (see Herve et al. (2004a) and Giepmans (2004) for review o f C x associating proteins). These associations are important not only for Cx43 trafficking, degradation, and function, but have also been implicated in cell adhesion and signaling. Given the plethora o f Cx43 associating proteins, many effects attributed to Cx43 may be mediated via these proteins and/or their interaction with Cx43 independent o f gap junction channel formation. While all o f these associating proteins are o f interest in their own right, several are o f particular interest to Cx43mediated motility.  1.6.2. Cadherins Cadherins are a widely expressed multigene family o f transmembrane proteins and are the major constituents o f adherens junctions. The extracellular N-terminus consists o f five tandem repeat domains which selectively mediate specific cell-cell adhesions via associations with other cadherins (Pecina-Slaus, 2003). The short, highly conserved cytosolic region associates with numerous proteins, most notably the actin cytoskeleton and actin binding  10 proteins including the a/p-catenins, vinculin, a-actinin, and zona occludens-1, and determines the specificity o f adhesive interactions (Klingelhofer et al., 2000; Nagafuchi, 2001). The extracellular and cytoplasmic domains enable cadherins to act as both receptors and ligands. Cadherins are divided into subclasses (including the predominant E , N , and P subclasses) depending on tissue distribution (Takeichi, 1988). The function and localization o f cadherins to cell-cell contacts may be a prerequisite for gap junction formation as disruptions in cadherin function can perturb gap junction formation and impair or inhibit G J I C (Frenzel and Johnson, 1996). Furthermore, C a -dependent regulation o f GJIC involves C a -dependent cell adhesion molecules (i.e. E-Cadherin (Jongen et al., 1991; Wang and Rose, 1997). N - C a d h e r i n N-Cadherin (N-Cad, Neural cadherin) is a 130kDa protein and it is the major adhesion protein in the C N S (Hatta et al., 1985). While the loss o f E-Cadherin is associated with endothelial to mesenchymal transition ( E M T ; Takeichi, 1988), N - C a d expression is often associated with increased invasion (Mareel and Leroy 2003). A potential mechanism involves the interaction o f N - C a d with F G F R - 1 which leads to prolonged mitogen activated protein kinase ( M A P K ) / E R K (extracellular signal regulated protein kinase) activity and increased matrix metalloprotease-9 ( M M P - 9 ) gene transcription (Wheelock and Johnson, 2003). N - C a d has been reported to associate with other proteins including actin, catenins, and C x 4 3 . The N-Cad/Cx43 interaction has been studied extensively in neural crest ( N C ) and N I H 3 T 3 cells. Knockdown o f N-cad expression by s i R N A in N I H 3 T 3 cells decreased the cell surface expression o f Cx43, and similarly the knockdown o f Cx43 by s i R N A reduced the cell surface expression o f N - C a d in these cells (Wei et al., 2005). The C x 4 3 / N - C a d interaction also appears to be essential for the function o f gap junctions containing Cx43 as N-Cad-deficient neural crest cells exhibit reduced GJIC despite retaining abundant gap junction plaques containing Cx43 ( X u et al., 2001). The association o f these proteins with one another has particularly important physiological  11 consequences as perturbations in their expression and/or localization modulate cell motility (Wei et al., 2005; X u et al., 2001). The specific site o f the N - C a d / C x 4 3 interaction is not yet known (Wei et al., 2005).  1.6.3 Catenins Catenins are a family o f proteins best known for their cytoplasmic association with cadherins. This association serves to help cluster cadherins and produces a complex which is linked to the actin cytoskeleton (Aberle et al., 1996; X u et al., 2004). Thereby, catenin-cadherin associations play an important role in cadherin-mediated cell adhesion and cytoplasmic functions o f cadherins (Gumbiner and M c C r e a , 1993). Importantly, catenins are also involved in biochemical pathways and are able to mediate transcription. P120 catenin p i 2 0 catenin (pl20ctn), a 120kDa protein associates with the cytoplasmic domain o f N Cad, facilitating N - C a d clustering (Wei et al., 2005) and has also been shown to colocalize with Cx43 ( X u et al., 2001). Cells in which Cx43 has been knocked down by s i R N A exhibit a reduction in the cell surface expression o f pl20ctn and an increase cytoplasmic and nuclear p l 2 0 c t n were observed ( W e i et al., 2005). The cytoplasmic and nuclear localization o f pl20ctn is associated with increased malignancy (Bellovin et al., 2005; van R o y and M c C r e a , 2005). Cytoplasmic p l 2 0 c t n inhibits Rho kinase and activates Rac and Cdc42 (Anastasiadis et al., 2000; Anastasiadis and Reynolds, 2001; Grosheva et al., 2001; H o u et al., 2006) and the nuclear localization o f p l 2 0 c t n inhibits kaiso-mediated transcriptional repression (van R o y and M c C r e a , 2005). Therefore, p l 2 0 c t n signaling has been proposed to mediate the cross-talk between Cx43 and cell motility mechanisms ( X u et al., 2001).  1.6.4 Drebrin Drebrin is a soluble 70 k D a protein that binds and stabilizes F-actin (Asada et al., 1994). In addition to localizing to plaques o f adhering junctions, drebrin localizes to leading edges o f  12 lamellipodia and filopodia and plays a role to the formation o f protrusions, thereby contributing to cell motility (Peitsch et al., 2001). Drebrin was identified as an interaction partner o f the Cx43 C-terminal domain at the plasma membrane in astrocytes and is required to maintain functional Cx43-containing gap junctions at the cell surface (Butkevich et al., 2004). 1.6.5 N O V (Nephroblastoma Overexpressed) A l s o known as C C N 3 , N O V was originally identified as an aberrantly expressed gene in avian nephroblastomas induced by myeloblastosis-associated virus ( L i n et al., 2003a). It is a member o f the Cysteine-Rich61/Connective Tissue Growth Factor/Nephroblastoma Overexpressed ( C C N ) family o f genes. It is upregulated in Cx43 transfected cells ( M c L e o d et al., 2001) and has been shown previously to associate with the C-terminal tail o f Cx43 (Fu et al., 2004; Gellhaus et al., 2004). Consistent with a growth suppressive role for C x 4 3 , N O V inhibits glioma cell growth and tumorigenic potential (Gupta et al., 2001). N O V induces directed endothelial cell migration ( L i n et al., 2003a), promotes migration and invasion in Ewing's sarcoma (Benini et al., 2005), and leads to an increase in glioblastoma cell migration v i a a platelet-derived growth factor receptor (PDGFR)-alpha-dependent mechanism (Laurent et al., 2003). 1.7 Involvement of C x 4 3 i n Motility While the anti-proliferative effects o f Cx43 are well established, its role in motility remains unclear. In skin wound healing assays, a loss o f Cx43 staining at the wound margins during initial wound healing and persistence o f Cx43 expression in non-healing wounds (Brandner et al., 2004) supported the finding by (Qiu et al., 2003) that targeting Cx43 expression by antisense technology enhanced the rate o f wound repair. However, no correlation between Cx43 expression and cell migration or invasion was observed upon C x 4 3 transfection into in basaloid squamous cell carcinoma (Shima et al., 2006). K n o c k d o w n o f Cx43 expression by s i R N A led to enhanced invasion o f M D A - M B - 2 3 1 and Hs578T breast carcinoma cells (Shao et  13 al., 2005). Complete ablation o f Cx43 by knockout methods were employed for explant outgrowth studies by L o and coworkers. In proepicardial explants from C x 4 3 K O mice outgrowth was enhanced ( L i et al., 2002). In contrast, neural tube explants from C x 4 3 K O mice exhibited decreased neural crest cell outgrowth (Huang et al., 1998a). Supporting these findings, increased outgrowth o f neural crest cells was observed in neural tube explants o f C M V 4 3 transgenic mice (animals in which Cx43 is overexpressed by the cytomegalovirus ( C M V ) promoter; Huang et al., 1998a). Furthermore, Cx43 knockdown by s i R N A inhibited NLH3T3 cell motility (Wei et al., 2005). This motility promoting ability o f Cx43 has also been observed in several overexpression studies. For example, stable transfection o f Cx43 into H e L a (Graeber and Hulser, 1998) and C 6 cells ( L i n et al., 2003b; Oliveira et al., 2005; Zhang et al., 2003b) increased cellular motility and invasion. Finally, examination o f the human glioblastoma cell lines, G L 1 5 and 8 - M G , human glioma biopsies xenografted and maintained i n nude mice, and fresh human glioma biopsies revealed that endogenous Cx43 expression was proportional to cellular motility and invasion (Oliveira et al., 2005)  1.7.1 Mechanisms of Cx43 mediated motility Although the mechanism by which Cx43 influences motility, irrespective o f affect, has not yet been determined, junction-dependent and junction-independent mechanisms have earned nominations. Junction dependent mechanisms have traditionally explained the effects o f Cxs and it has been suggested that in the context o f cancer invasion this might owe to the ability o f cancer cells to metabolically hijack adjacent cells (Wei et al., 2005). Junction independent mechanisms, however, are equally able to explain such phenomena. The disruption o f cadherin localization and adherens junction formation upon Cx43 knockdown ( W e i et al., 2005) along with the finding that deletion o f the Cx43 carboxy terminus attenuates motility o f 3T3 A 3 1 fibroblasts (Moorby et al., 2000) supports junction-independent mechanisms. Furthermore, using a gene array approach,  14 Iacobas et al. (2004) have identified 18 genes involved in motility that are differentially regulated in C x 4 3 K O astrocytes compared to w i l d type. 1.8 C o n n e x i n 4 3 i n cancer The implication o f gap junctions in cancer was first described by Loewenstein and Kanno in 1966 when they noted that G J I C was decreased in liver tumor cells compared to untransformed counterparts. Cx43 has since been implied in carcinogenesis, tumor susceptibility and progression, as well as tumor suppression and differentiation. The reduction or elimination o f Cx43 in several types o f cancer has been documented (see above). Reconstitution o f Cx43 in cells derived from these tumors has reduced their proliferation and/or tumorgenicity and directly identified Cx43 as a tumor suppressor (Fernstrom et al., 2002; Fishman et al., 1990; Goldberg et al., 2000; Jou et al., 1993; Naus et al., 1992; Shima et al., 2006; Z h u et al., 1992). There also exists considerable indirect evidence for the tumor suppressive role o f Cx43. Compounds involved in decreasing tumorigenicity such as retinoids, carotenoids (Bertram and Vine, 2005), vitamin A (Naves et al., 2001), D i a l l y l disulfide ( D A D S ; Huard et al., 2004), and C N T F (Ozog et al., 2002; Ozog et al., 2004), have led to parallel increases in C x 4 3 . Furthermore, the expression o f tumor suppressors (e.g. breast cancer metastasis suppressor 1 ( B R M S 1 ; Kapoor et al., 2004) correlates with increased levels o f C x 4 3 . Complimentary studies have investigated the correlation between the loss o f Cx43 and increased tumorigenicity. Numerous carcinogens result in decreased Cx43 (reviewed by (Trosko and Ruch, 2002) including the phorbol ester tumor promoter 12-O-tetradecanoylphorbol 13acetate ( T P A ; Enomoto et al., 1981; Swierenga and Yamasaki, 1992), A^AT-nitroso-methylurea (Benda et al., 1968), and dimethylbenzanthracene ( D B M A ; Kamibayashi et al., 1995). Similarly, the overexpression o f oncogenes that increase tumorigenicity including Src, N e u , Ras, and M y c result in a downregulation o f Cx43 protein or function (Azarnia et al., 1989; K a l i m i et al., 1992;  15 Jou et al., 1995; Hayashi et al., 1998). Additionally, heterozygous (Cx43(+/-)) mice exhibit reduced expression o f Cx43 and have an increased susceptibility to tumorigenesis upon exposure to environmental agents such as urethane (Avanzo et al., 2004). 1.9 B r a i n cancer The most comprehensive study o f brain cancers has been undertaken by the Central Brain Tumor Registry o f the United States ( C B T R U S ) and is therefore the source o f the statistics provided in this section (unless indicated otherwise). L i k e most cancers, brain cancers can be malignant or benign however benign brain cancers are not necessarily harmless -they can be life threatening i f they compromise vital brain activity (by occlusion, edema, or otherwise) and lead to neurological dysfunction. Indeed prognosis is dependent on tumor location: for malignant brain tumors the five year survival rates differ markedly between tumors located in the frontal, temporal, parietal, occipital lobes o f the brain compared to tumors located i n the brain stem, pituitary gland, pineal gland, or cerebellum ( C B T R U S ) . Brain cancers are rarely metastatic but can originate from within the brain de novo (primary), which is frequently associated with epidermal growth factor receptor ( E G F R ) amplification, or arise as a result o f metastasis from elsewhere (secondary), which is associated with tumor protein53 (TP53) inactivation (Benjamin et al., 2003). 1.9.1 G l i o m a Gliomas are the most common o f malignant central nervous system ( C N S ) cancers accounting for greater than 78% o f primary malignant C N S tumors (including tumors o f the spinal cord, C B T R U S ) . G l i o m a is a category o f primary brain cancer, specifically a type o f neuroepithelial tumor, which arises from the neoplastic transformation o f glial cells, or perhaps their stem cell precursors (Singh et al., 2003). The neoplastic transformation o f oligodendrocytes (or their precursors) accounts for 8% o f gliomas while the neoplastic transformation o f astrocytes accounts for 69% o f gliomas ( C B T R U S ) .  16 Biopsied gliomas are divided into categories according to presumed cellular origin; astrocytic, oligodendroglial, mixed gliomas (oligoastrocytoma and anaplastic oligoastrocytoma), ependymal tumors, and neuroepithelial tumors o f uncertain origin. These categories are further graded according to the world health organization ( W H O ) classification system (which is the most widely employed o f such systems, see Louis et al., 2001). Astrocytomas, the most prevalent o f gliomas, are then histologically tiered into four grades according to their malignancy (in increasing order o f malignancy): grade I (pilocytic astrocytomas), grade II (low-grade nonpilocytic/diffuse astrocytomas), grade III (anaplastic astrocytomas), and grade I V (glioblastoma multiforme ( G B M ; Kleihues P, 2000). A comprehensive review o f cases reported between 1972 and 2002 indicates that the relative survival rates o f patients differ markedly between tumor grades: the 10 year survival rate for pilocytic astrocytoma is 89.3% while the 2 year survival rate for glioblastoma multiforme is 8.7% ( C B T R U S ) . Unfortunately, glioblastoma is the most common o f gliomas accounting for nearly 50% o f gliomas ( C B T R U S ) . 1.9.2 N a t u r e of gliomas Glioblastomas form aggressive tumors which grow quickly and infiltrate relentlessly. The tumors spread through the brain along surfaces and white matter tracts and deccusation v i a the corpus callosum and invasion into the contralateral hemisphere is a common characteristic o f these neoplasms. Furthermore, in vitro assays using spheroids o f rat glioma (C6) cells provide evidence for the secretion o f a chemorepellent by the tumor cell mass that promotes the invasion of glioma cells (Werbowetski et al., 2004). In these studies, cells responded to a gradient o f chemorepellent cues and migrated out o f the spheroid at an angle perpendicular to the spheroid edge. The addition o f an adjacent spheroid altered the trajectory o f cell motility away from both spheroids while cells readily invaded adjacent astrocyte aggregates (Werbowetski et al., 2004).  17 According to mathematical modeling, the diameter o f glioblastoma expands at a rate o f 2.5mm/month which is much more explosive than the expansion rate for lower grade astrocytomas (4.4mm/year, Swanson et al., 2003). 1.9.3 Causes  The W H O grading system is very valuable as it not only dictates prognosis and treatment strategies but it also provides a framework for research as excised tissue is examined for morphological and biochemical changes (Louis et al., 2001). W h i l e the causes o f gliomas remain unknown, important advancements in both detection and treatment have been accomplished in this manner, leading to the identification, classification, and expression profiling o f several important predisposing genes. TP53 mutations, often associated with gliomas, are significantly more frequent in lowgrade astrocytomas suggesting that they may contribute to the initiation o f these tumors (Ohgaki and Kleihues, 2005a). The most common inherited tumor predisposition syndromes in the C N S arise due to mutations in the tumor suppressor neurofibromatosis 1/2 (NF1/2) genes which cause neurofibromatosis 1/2 and render patients with an increased susceptibility to astrocytoma (Korones et al., 2003; Gutmann and Giovannini, 2002). Alterations in papillary thyroid carcinoma ( P T C ) and adenomatous polyposis coli ( A P C ) genes are the inherited source o f Turcot syndrome and Gorlin syndrome respectively and are associated with medullablastoma (Gutmann and Giovannini, 2002), the most common primary brain tumor found in children ( C B T R U S ) . Patients who have germline mutations in the tuberous sclerosis complex-2 (TSC2) gene acquire tuberous sclerosis and have an increased susceptibility to subependymal giant cell astrocytomas (Woods et al., 2002). O f note, several studies have implicated Cx43 in glioma progression. W h i l e astrocytes normally express very high levels o f Cx43, its expression is decreased i n low grade tumors and it is often absent in glioblastoma (Soroceanu et al., 2001; Huang et al., 1999; P u et al., 2004).  18 Environmental factors may well play a role in initiation or progression o f these tumors. While exposure to vinyl chloride has been associated with these neoplasms (Moss, 1985), only therapeutic X-irradiation is unequivocally associated with an increased risk o f brain tumors, including gliomas (Ohgaki and Kleihues, 2005a) 1.9.4 Treatment/management strategies Although surgical resection o f the tumor mass has provided invaluable insights for research and treatment and is important for reducing cranial pressure, the invasive nature o f diffuse gliomas often precludes surgery as an effective means to cure the disease. Similarly, chemotherapy and radiotherapy are not curative as these approaches target only dividing cells. Because division and invasion are mutually exclusive events (Chicoine and Silbergeld, 1995), non-dividing cancerous cells are able to reenter mitosis at a later time. Combination therapy approaches involving resection, chemotherapy and radiotherapy can best be described as management strategies for the betterment o f quality o f life -although the incidence o f glioblastoma is low (3.05 per 100,000, C B T R U S ) , so too is the median survival time o f 0.4 years (Ohgaki and Kleihues, 2005b). Gene therapy is hence an attractive therapeutic approach. The transfer o f the herpes simplex thymidine kinase ( H S V t k ) gene to neoplastic cells renders their viability sensitive to ganciclovir ( G C V ) , an antiherpetic drug. The toxic metabolites produced upon H S V t k phosphorylation o f G C V can be transferred via gap junctions to adjacent cells (Elshami et al., 1996). This process, called the 'bystander effect', leads to the killing o f untransduced cells adjacent to cells expressing the H S V t k gene (Elshami et al., 1996). The first in vivo study involving Cx43 and the H S V t k - G C V approach measured tumor formation when mixed cultures o f transduced and non-transduced C 6 cells were injected subcutaneously into CB.17/SCID-beige mice (Dilber et al., 1997). When C 6 cells were transfected with Cx43 and exhibited a high level o f G J I C , tumors were frequently undetectable at the inoculation site, even when 75% o f cells were H S V t k negative. Aggregates o f cells from primary astrocytomas,  19 primary astrocytoma cell cultures, and glioblastoma cell lines demonstrated a varied bystander effect during H S V t k gene therapy depending on the level o f Cx43 m R N A expression (Shinoura et al., 1996). Unfortunately, the H S V t k - G C V approach has not yet proven successful for gliomas due to low sensitivity o f these tumor cells to viral infection and further by a limited bystander effect due to low levels o f Cx43 (Rosolen et al., 1998).  1.10 C6 Cells The C 6 cell line was originally isolated from tumors in Wistar-Furth rats (rattus norvegius) exposed to N,N'-nitroso-methylurea  and based primarily on histopathological features  C 6 is categorized as astrocytoma (Benda et al., 1968). These cells model human gliomas, particularly G B M , in that they are morphologically similar to human gliomas and exhibit diffuse invasion when injected into the brains o f healthy neonatal rats (Grobben, A u e r et al 1981). Although when C 6 spheroids are implanted into neuronal tissue circumscribed tumors are formed (Naus et al., 1992), when implanted in suspension the cells invade individually as single cells located in micropockets (Bernstein et al. 1990, Izumoto et al. 1996, Goldberg et al. 1991, Grobben et al., 2002). C 6 cells therefore provide an adequate means to study biochemical pathways, growth, invasion, and angiogenesis o f glioblastomas both in vitro and in vivo. In addition, it is an ideal system to examine Cx43 as similar to human tumor tissue, Cx43 is downregulated. The persistence o f Cx43 in this cell line ensures that cells express the appropriate machinery required for the proper gating, localization and function o f C x 4 3 . Furthermore, the cells also form confluent monolayers in vitro and are therefore amenable to wound healing assays unlike other glioma cell lines such as the murine delayed brain tumor ( D B T ) and human U87.  20  1.11 Rationale Various genetic and biochemical changes occur during the neoplastic transformation of astrocytes. There is a downregulation of Cx43 expression throughout glioma progression and expression is often absent in glioblastoma multiforme. Several studies have indicated that Cx43 enhances motility in glioma cells, however all have taken an exogenous overexpression approach. It has been reported that exogenous Cx43 overexpression affects Cx43 trafficking. Furthermore, Cx43 transfected cells have a different gene expression profile than untransfected cells. Here we aim to investigate the role of endogenous Cx43 in glioma motility and further explore the mechanism by which such a role is carried out. We hypothesize that the Cx43 C-terminus may play a significant role for Cx43-mediated motility in C6 cells.  21 F i g u r e 1.1 A c t i n P o l y m e r i z a t i o n . Activated Rac and the adaptor protein N e k cause the dissociation o f S c a r / W A V E from a complex composed o f H S P C 3 0 0 , A b i , Sra-1, and Nap 125 while activated cdc42 activates W A S P and N - W A S P . W A V E / S c a r - H S P C 3 0 0 and W A S P / N - W A S P both activate the A R P 2 / 3 complex which binds to the pointed end o f F-actin and causes extension o f nascent filamentous (F)-actin at a 70° angle. Activated cdc42 also activates mDia2 which, when associated with monomeric, globular (G)-actin bound profilin, binds to the pointed end o f F-actin and causes extension o f F actin in a linear fashion. R h o A results in the activation o f the m D i a l which, when associated with G-actin bound profilin, also binds to the pointed end o f F-actin and causes extension o f F actin in a linear fashion. The interaction o f V A S P with vinculin is also important for actinfilament assembly. Both Rac and R h o A activate L I M K which in turn phosphorylates and inactivates cofilin preventing the disassembly o f F-actin into G-actin. E n a / V A S P associated with profilin interact with Z y m i n and a-actinin to stabilize the barbed end o f F-actin as does the capping protein gelsolin, which is under the control o f phosphoinositides.  22  C^^filirT^)  CEnAA/ASP>  ^<<«««^  )rofilin  phosphoinositides  gelsolinJ:«««  (^Rac)  (cdc42)  RhoA  23 Figure 1.2 Connexins, Connexons, and Gap Junction Formation. A ) Connexin topology. Connexins have 4 highly conserved transmembrane domains, two extracellular loop domains and one cytosolic loop domain, and both N and C termini are cytosolic facing. B ) Connexon assembly. The noncovalent hexameric oligomerization o f connexins forms a connexon or hemichannel. C ) Gap Junction assembly. Connexons from adjacent cells associate with one another to form transmembrane channels. Clustering o f these channels forms plaques which are called gap junctions. These junctions connect the cytosol o f adjacent cells allowing the selective and regulated movement/exchange o f amino acids, ions, small molecules and metabolites (<lkDa) between cells and also electrically couples adjoining cells.  Extracellular space  C a , IP , K , A T P 2 +  +  3  Table 1. Implication of Gap Junctions in Disease Disease State  Connexin involvement  Reference(s)  Epilepsy/epileptogenesis  Associated with alterations in G J I C Associated with mutations in G J B 1 encoding Cx32 Associated with mutations in G J B 2 encoding Cx26 Associated with mutations in G J B 2 encoding Cx26 Caused by mutations in G J A 1 encoding Cx43  Perez Velazquez and Carlen, 2000 Bergoffen et al., 1993  Charcot Marie Tooth Disease (X-linked) Deafness (nonsyndromic) sensorineural Keratitis-IchthyosisDeafness Syndrome Oculodentodigital Dysplasia  Kelsell etal., 1997  Richard et al., 2002  Paznekas et al., 2003  26  F i g u r e 1.3 Types of G a p J u n c t i o n Channels. Several types o f gap junction channels can be classified based on the compliment and organization o f connexins which compose the connexon hemichannels. When a connexon is composed o f only one type o f connexin, it is called a homomeric hemichannel. The association o f identical homomeric hemichannels creates homomeric homotypic gap junction channels while the association o f non-identical homomeric hemichannels creases a homomeric heterotypic channel. Connexons composed o f more than one type o f connexin are called heteromeric hemichannels. The association o f identical heteromeric hemichannels (i.e. containing the same compliment o f connexins) creates heteromeric homotypic gap junction channels while the association o f non-identical heteromeric connexons creates heteromeric heterotypic gap junction channels.  27  r o  X CD  CD  c c o O  c c  CO  sz O  Connexon  Homomeric  Homomeric  Heteromeric  Heteromeric  Channel  Homotypic  Heterotypic  Homotypic  Heterotypic  28 CHAPTER 2 MATERIALS AND METHODS 2.1 Cell Culture C6 rat astrocytoma cells (Benda & Lightbody 1968) were grown on non-pyrogenic, N u n c l o n ™ A Surface plates ( N U N C ) and maintained in low glucose D M E M containing L-glutamine (Sigma) supplemented with 10% Fetal Bovine Serum (FBS, Sigma), 100 units/ml penicillin, and 100 ug/ml streptomycin in a humidified incubator (37°C; 5% C 0 2 / 9 5 % air). H E K 2 9 3 retroviral packaging cells (generous gift from Dr. Richard C . Mulligan, Children's Hospital, Boston, M A ) were maintained in low glucose D M E M containing L-glutamine (Sigma) supplemented with 10% F B S (Sigma), 1 ug/ml tetracycline, 2 ug/ml puromycin, 0.3 ug/ml G418, and 100 units/ml penicillin, and 100 ug/ml streptomycin i n a humidified incubator (37°C; 5% C 0 2 / 9 5 % air). See Transfection  and infection of shRNA  (below) for details regarding  generation o f infectious medium. 2.2 Gap junction blocker treatments The non-selective gap junction blocker carbenoxolone ( C B X , Sigma) was dissolved in sterile H 2 O to final concentration o f 150uM. The inactive analogue, glycyrrhizic acid ( G Z A , Sigma) was also dissolved in sterile H 2 O to a final concentration o f 1 5 0 u M and used as a comparative control. Based on previous studies (Bani-Yaghoub et al., 1999) and preliminary tests, the lowest concentration o f C B X that completely blocked the transfer o f calcein (150uM) was used for experiments in this study. 2.3 Immunocvtochemistry Cells grown on coverslips were fixed in 4% paraformaldehyde for 10 min at 37°C and were subsequently washed twice at 37°C for 10 min in phosphate buffered saline (PBS). Coverslips were incubated in blocking solution (4% normal goat serum or bovine serum albumin) for 30 min at room temperature followed by incubation for l h at room temperature in blocking solution containing primary antibody against Cx43 (Sigma, rabbit polyclonal, 1:400), N - C a d (mouse  29 monoclonal, B D Biosciences, 1:500), mouse monoclonal anti-pl20ctn (mouse monoclonal, B D Biosciences, 1:500),. Coverslips were then washed three times for 10 m i n in P B S prior to incubation for l h at room temperature in the dark in blocking solution containing secondary antibody (Molecular Probes, goat anti-rabbit IgG, 1:100). Coverslips were again washed three times for 10 m i n in P B S at room temperature in the dark and then rinsed in ddH^O. Coverslips were mounted onto microscope slides with Prolong G o l d ® containing 4',6-Diamidino-2phenylindole ( D A P I , Molecular Probes). Note that no immunoreactivity could be detected when the Cx43 antibody was used in experiments performed on astrocytes taken from C x 4 3 K O mice. 2.4 Fluorescence Microscopy (Image acquisition) Fluorescent and differential interference contrast (DIC) or brightfield images were acquired with an A x i o C a m M R m (Zeiss) using the Axioskop2 epifluorescent microscope (Zeiss). Between cell types and treatments, the same exposure time was used and any modifications (i.e. min/max) were performed equally using A x i o V i s i o n software (Zeiss). 2.5 Protein Collection Cells were grown on 60mm dishes. Upon reaching confluency, cells were rinsed in ice cold P B S and lysed i n 250ul radioimmune precipitation lysis buffer ( R I P A ; 0.005% sarkosyl w/v, 0.01% I G E P A L v/v, 0.001% S D S w/v, 1 5 0 m M N a C l , 50mMTris, pH8.0) buffer containing phosphatase (Roche) and protease inhibitors (minicomplete, Roche) until lysis was complete. After D N A was sheared by titration with a fine gauge needle or sonicator, samples were centrifuged at 4°C for 10 m i n at 14,000rpm. The protein-containing fraction (supernatant) was subjected to the colorimetric B C A Protein Assay K i t (Pierce). Results were read at a wavelength o f 570 nm on a W a l l a c l 4 0 0 plate spectrophotometer (Wallac) to determine protein concentration.  30 2.6 S D S - P A G E  Protein samples were diluted in R I P A and denaturing sample buffer containing O . l u l / m L (3mercaptoethanol. After boiling the samples for 3min, a volume containing 30ug was loaded onto 8% (for N-Cadherin and pl20ctn) and 10% (for N O V , C x 4 3 , and G A P D H ) acrylamidecontaining gels. Electrophoretic separation was conducted in a pre-mixed buffer (0.3% w/v Tris, 1.44 % glycine w/v, 1% S D S w/v) using a Mini-Protean 3 electrophoresis system (Biorad) at 100V for approximately 2 hours. Separated protein was electrically transferred (15V, 35min) onto an I m m u n - B l o t ® P V D F (polyvinylidene difluoride) membrane (Biorad) using a Transblot S D ® semidry transfer cell (Biorad). 2.7 Immunoblotting Blots were equilibrated for approximately 5 min in Tris-Buffered Saline ( T B S ; 5 0 m M Tris, 150mM N a C l , pH7.5) following semidry transfer and subsequently incubated in a solution of 5% non-fat dried milk (Carnation, Nestle) in T B S containing 0.1% Tween ( T B S T ) for 1 hour at room temperature to block non-specific antibody binding. Blots were then incubated in a 1% m i l k - T B S T solution containing antibodies to the protein o f interest for l - 2 h at room temperature or overnight at 4°C. Antibodies included: rabbit polyclonal anti-Cx43 (Sigma, 1:8,000), rabbit polyclonal a n t i - N O V (Fu et al. 2004; 1:800), mouse monoclonal anti-GFP (Chemicon, 1:1,000), mouse monoclonal anti-N-Cadherin ( B D Biosciences, 1:2500), mouse monoclonal antip i 20Catenin ( B D Biosciences, 1:1,000), and mouse monoclonal a n t i - G A P D H (HyTest Ltd., 1:10,000). Blots were then washed with 3 changes o f T B S T for 10 minutes each at room temperature and incubated in a l % m i l k - T B S T solution containing appropriate secondary antibody conjugated to horseradish peroxidase (goat anti-rabbit H R P or goat anti-mouse H R P , Ceaderlane, 1:10,000) for l h at room temperature. Finally, blots were washed at room temperature with 2 changes o f T B S T for 10 minutes each, followed by an additional 10 min  31 wash with T B S and incubated for 5 minutes at room temperature in SuperSignal chemiluminescent substrate (Pierce) before exposure to X - r a y film (Kodak). Densitometry analysis was carried out using A l p h a l m a g e r ™ 3 4 0 0 software (Alphalnnotech) to determine relative changes in protein levels. 2.8 Preloading/Dye Coupling Intercellular coupling between cells was evaluated according to the established preloading method described by Goldberg et al.(1995). Briefly, cells were seeded in 10% F B S containing medium into 35-mm plates. Upon reaching confluency, donor cells were incubated in a dye solution (5 u M c a l c e i n - A M [Molecular Probes] and 10 u M D i l [Sigma-Aldrich] in an isotonic [0.3 M ] glucose solution) for 20 min in an incubator. Donor cells were then rinsed several times with isotonic glucose, trypsinized, seeded onto recipient sister cultures at a ratio o f 1:500, and maintained for 3.5 h in the incubator. Cells were examined by epifluorescence microscopy and gap junctional coupling was assessed by the passage o f calcein from donor cells to, and among, recipient cells. 2.9 Wound healing motility assay The following assay was modified from Lagana and Goetz et al (2006). Cells (1-2 x l O ) were 6  grown to confluency (l-3d) on 60mm dishes in D M E M supplemented with 10%FBS and penstrep. The confluent monolayer was scraped using a cell lifter (Costar) i n close proximity to a line which had been etched onto the underside o f the dish using a razorblade. Cells were rinsed several times in P B S to remove loose and detached cells. Images were acquired immediately and 24h after the scrape. The images acquired at time 0 and time 24h were overlain in Adobe Photoshop. Migration distances were measured using A x i o v i s i o n software (Zeiss) by drawing a line from the edge o f the scrape at time 0 to the leading edge o f the ten most migratory cells after 24h. (Unpaired t-tests and one way A N O V A were preformed using Instat software, see statistical analysis).  32 2.10 Transfection and infection of shRNA Transfections and infections were modified from Shao et al. (2005) and M a o et al. (2000) respectively. R N A i - 1 and R N A i - 2 s h R N A sequences in retroviral p H l . l Q C X I H (GeneScript) vectors were transfected into H E K 2 9 3 packaging cells using L i p o f e c t A M I N E 2000 reagent and M E M . (Invitrogen, 20iig plasmid onto 100mm dishes). 6 hours following transfection 3 m L o f serum-free D M E M containing tetracycline was added. 24h post transfection the medium was replaced with serum-containing D M E M without tetracycline to allow for the production o f the G glycoprotein o f vesicular stomatitis virus ( V S V G ) . 48-72h following transfection the medium was removed and filtered (0.45 um filter, B D Biosciences). C e l l culture media o f C 6 - H cells was replaced with filtered retroviral supernatant for infection. Following three rounds o f infection cells were further cultured in selection media containing 100(xg/ml hygromycin. Antibioticresistant cells were passaged in hygromycin-deficient medium prior to further experimentation. The entire cell population was kept rather than isolating clones. 2.11 Transwell migration invasion assays Transwell assays were modified from (Zhang et al., 2003a; Vigetti et al., 2006; and www.fisher.co.ukytechzone/life/table-pdfs/protocol/transwell_trypsin_protocol.pdf). M e d i u m (600ul) was added to lower compartment and 7 . 5 x l 0 cells (suspended i n 200ul 4  medium) were added to upper chamber. After 10-14h incubation at 37°C inserts were removed and transferred to 24 well plates in which each well contained 600ul 0.25%Trypsin-EDTA. Following a 10 min incubation at 37°C the 24 well plate was agitated gently by vortexing to detach cells from the underside o f the P E T (polyethylene terephthalate) membrane. 500ul o f cell suspension was counted in a C O U L T E R C O U N T E R ® Z l Series Particle Counter (Coulter Electronics, Burlington, Ontario). ( N O T E : For invasion experiments, 200ul 5% gelatin (in P B S ) was added to upper chamber and allowed to dry (24h). Prior to seeding cells, inserts were first rehydrated by adding 200ul medium to upper chamber.) Assays were performed in duplicate. To  33 determine the motility index, the number o f migrated cells was normalized to companion plate controls. 2.12 Statistical analysis Quantitative data were applied to statistical analysis using InStat® software. Data were first subjected to a one-way analysis o f variance ( A N O V A ) where applicable followed by an unpaired, two-tailed Student's t-test to determine whether significant differences existed between samples.  34 C H A P T E R 3 THESIS R E S U L T S Section 3.1 Motility of C 6 - H subclones Similar to high grade human gliomas, C 6 rat astrocytoma cells typically express very low levels o f C x 4 3 . Previous studies investigating the consequence o f Cx43 on the motility o f C 6 cells have employed an exogenous overexpression approach ( L i n et al., 2002; Oliveira et al., 2005; Zhang et al., 2003a). Although these studies all indicate that Cx43 overexpression leads to an increased motility and/or invasion, it has been reported that such exogenous overexpression causes premature assembly and altered trafficking o f Cx43 (VanSlyke, 2005). Furthermore, we have previously shown that cells in which Cx43 is overexpressed have a very different gene profile than mock-transfected controls (Naus et al., 2000). W h i l e C 6 cells typically express low levels o f C x 4 3 , we have isolated subclones by serial dilution in which Cx43 is endogenously expressed at moderate to high levels ( C 6 - H , Figure 3.4A,B). Similarly, G J I C is also increased in these cells (Figure 3.4 C ) . W e exploited the C 6 - H subclones to examine the consequence o f endogenous upregulation o f Cx43 in C 6 cell motility. W o u n d healing assays (schematic, time Oh, 24h Figure 3.5A) were conducted to assess directional motility and it was found that the subclones endogenously expressing higher levels o f Cx43 traveled a distance significantly greater than parental C 6 cells expressing low levels o f Cx43 ( P O . 0 0 1 , Figure 3.6B). Transwell chambers (schematic, Figure 3.5B) were also used to measure the random (i.e. non-directional) motility o f these cells. Similar to directional motility, a significant (approximately 75%) reduction in the percentage o f randomly migrating cells was observed for C 6 - L cells compared to C 6 - H subclones (n=3, P O . 0 0 0 1 , Figure 3.7). Section 3.2 Knockdown of Cx43 in C 6 - H subclones by shRNA The results described in the previous section prompted us to investigate whether the increased motility exhibited by the C 6 - H cells could be attributed directly to increased levels o f  35 Cx43 or whether it was merely coincidental. W e employed an R N A i approach to explore this possibility. (C6-H) subclones were infected with Cx43 s h R N A or control (empty vector or scrambled Cx43 s h R N A ) constructs and the generation o f stably expressing cells (C6-Cx43shRNAempty, C6-Cx43shRNAscrambled, C 6 - C x 4 3 s h R N A - l , C 6 - C x 4 3 s h R N A - 2 ) was realized by hygromycin selection as described in the materials and methods. Consistent with previous observations (Shao et al., 2005), only the C x 4 3 s h R N A - l construct led to a reduction in Cx43 protein (Figures 3.8 and 3.9). Densitometric analysis revealed that in cells stably expressing the C x 4 3 s h R N A - l construct expression o f Cx43 was reduced to 36% o f scrambled control (Figure 3.8). To avoid clonal selection artifact and maintain the heterogeneity o f the C 6 cell line, the entire population of hygromycin-resistant cells was collected. Importantly, the high level o f Cx43 retained in the C6-Cx43shRNAscrambled cells, and the low level o f Cx43 in the C 6 - C x 4 3 s h R N A - 1 cells was uniform throughout the entire cell population as indicated by immunocytochemistry (Figure 3.9). Section 3.3 Knockdown of Cx43 by shRNA attenuates cell motility and invasion W o u n d healing assays were again conducted to examine the consequence o f Cx43 downregulation on directional cell motility. Untransfected, empty, and scrambled controls as well as cells expressing the C x 4 3 s h R N A - 2 construct (which did not knockdown Cx43 protein) migrated a distance significantly greater than cells in which the level o f Cx43 protein had been knocked down by the expression o f the C x 4 3 s h R N A - l construct (Figure 3.10). The motility o f cells expressing scrambled C x 4 3 s h R N A or C x 4 3 s h R N A - l was further investigated using transwell chambers to assess random (i.e. non-directional) motility. In this assay, the number o f migrating cells is examined rather than the migration distance (by counting the number o f cells that traverse a perforated (8um), uncoated P E T membrane, Figure 3.5B). In addition to a reduction in directional motility upon Cx43 knockdown, a 50% reduction in the  36 percentage of randomly migrating cells was observed for cells expressing the Cx43 shRNA-1 construct compared to cells expressing Cx43shRNA scrambled control (Figure 3.11). Notably, glioma cells do not simply migrate passively through the brain but also actively invade into the brain parenchyma. Therefore, the consequence of Cx43 knockdown on invasivity was also investigated using transwell chambers in which the perforated (8um) PET membrane had been coated with 5% gelatin prior to seeding the cells (Figure 3.5C). The consequence of Cx43 knockdown was more dramatic on cell invasivity than on cell motility: a 75% reduction in the percentage of invading cells was observed for cells expressing the Cx43 shRNA-1 construct compared to cells expressing Cx43shRNA scrambled control (Figure 3.12). Section 3.4 Mechanisms of Cx43 Mediated motility In addition to attenuated motility and invasivity, GJIC was also decreased in C6Cx43 shRNA-1 cells compared to C6-Cx43shRNAscrambled as indicated by restricted passage of calcein dye in preloading experiments (Fig 3.13). We reasoned that the attenuation in cell motility may be a direct result of attenuated GJIC. Therefore, i f functional gap junctions are essential for Cx43-mediated motility than blocking the channels would affect cell motility. Our next aim was consequently to elucidate a potential mechanism for Cx43 mediated motility. Previous studies have identified both junction-dependent (Oliveira et al., 2005) and -independent (Moorby, 2000; Wei et al., 2005) mechanisms. The non-selective gap junction blocker carbenoxolone (CBX) is widely employed for studying channel-dependent activity. In agreement with Bani-Yaghoub et al. (1999), application of 150uM C B X during preloading assays restricts dye passage to unlabeled recipient cultures while application of the same concentration of inactive analog, glycyrrhizic acid (GZA) does not restrict dye passage to adjacent unlabeled recipient cells (Figure 3.14, see materials and methods section for assay details).  37 To evaluate the contribution o f functional channel activity to C 6 motility we carried out directional and non-directional motility assays in the presence o f C B X . B l o c k i n g gap junction channel activity with C B X did not significantly attenuate the distance cells traveled in wound healing assays compared to untreated or control ( G Z A ) treated cultures (P>0.05, Figure 3.15). Similarly, the percentage o f cells that traversed through 8 urn pores on uncoated P E T membranes in transwell motility studies was not significantly increased upon treatment with 150 u M C B X compared to untreated or G Z A treated cells (n=5, Figure 3.16). Section 3.4.1 Protein localization and gap junctional intercellular communication in C6-Cx43ACT244-382GFP cells Because the results obtained in wound healing assays containing C B X indicated that there was no difference in directional motility, we considered a mechanism for Cx43 mediated motility involving the carboxy terminus o f C x 4 3 . T o this end, we employed C 6 cells which had previously been infected with either full length Cx43 tagged with G F P ( C 6 - C x 4 3 G F P ) or a truncated mutant in which the entire carboxy terminus had been deleted and replaced with G F P (C6-Cx43ACT244-382GFP; see Figure 3.17 and (Fu et al., 2004). The truncated protein localized to the periphery o f the plasma membrane and formed obvious gap junction plaques at sites o f cell-cell contact (Figure 3.18A). Immunocytochemistry revealed that C 6 - C x 4 3 A C T 2 4 4 - 3 8 2 G F P cells formed plaques composed entirely o f truncated Cx43 protein (Figure 3.18B, green punctate staining) in addition to heteromeric plaques composed o f both truncated Cx43 protein and endogenous Cx43 protein (Figure 3.18B, yellow punctate staining). Because heteromeric plaques composed o f both truncated Cx43 protein and endogenous Cx43 protein were abundant, there existed the possibility that the truncated Cx43 protein eliminated the ability o f endogenous Cx43 to form functional channels. W e therefore conducted dye preloading assays to determine whether the C 6 C x 4 3 A C T 2 4 4 - 3 8 2 G F P cells were communication competent. D y e passage was observed in these  38 cells (Figure 3.19) eliminating the confounding variable that carboxy truncation o f Cx43 may have imposed. 3.4.2 Expression of full length Cx43 enhances directional motility The consequence o f Cx43 carboxy terminus deletion for cell motility was first examined using wound healing assays. W h i l e exogenous expression o f full length C x 4 3 - G F P demonstrated increased motility compared to control ( G F P only), the communication-competent C 6 C x 4 3 A C T 2 4 4 - 3 8 2 G F P cells did not exhibit differences in motility compared to control cells (Figure 3.20). N o differences in the non-directional motility o f these cells was detected when assessed using uncoated transwell chambers in the absence o f an exogenous stimulus. (Figure 3.21, n=5 P>0.05). 3.4.3 Cx43 associating proteins in cells expressing Cx43A244-382GFP A mechanism involving the carboxy terminus o f Cx43 for Cx43-mediated motility was implied in the rate o f directional motility and furthermore we observed that cells expressing truncated protein appeared to migrate in collective sheets while control cells and cells expressing full length Cx43 migrated as individual cells in wound healing assays (Figure 3.22). Therefore, the expression and localization o f Cx43 associating proteins with established roles in motility were examined. N-Cadherin levels were similar between C 6 - G F P , C 6 - C x 4 3 G F P , and C 6 Cx43A244-382GFP cells (Figure 3.23A). Furthermore, N-Cadherin localized to the membrane and at regions o f cell-cell contract i n both control and C6-Cx43A244-382GFP cells (Figure 3.23, panel B (compare upper and lower). Similarly, neither protein levels nor localization o f pl20ctn were obviously different between these cells (Figure 3.24, panels A and B ) .  Figure  3.4 Connexin43 expression in C6 subclones. C 6 subclones generated by serial dilution  gave rise to populations o f cells in which Cx43 expression was enhanced (C6-H) compared to parental C 6 cells (C6-L) as detected by A ) Western immunoblotting with G A P D H serving as a loading control and B ) immunocytochemistry (Bar=20um). C ) D y e preloading revealed that C 6 H cells exhibit a greater degree o f coupling than C 6 - L cells (Bar=50um).  40  A  C6-H  C6-L  kDa  Cx43  43  37 GAPDH  B  C6-L  C6-H  * k  J • t vi • • —  42 Figure 3.5 Motility and Invasion assays. A ) Schematic of wound healing assays. Confluent monolayers were scraped in close proximity to a line which had been etched onto the underside of the dish using a razorblade. Images were acquired immediately (left) and 24h after the scrape (right) and were subsequently overlaid in Adobe Photoshop ® (lower) to accurately determine the site of the lesion (arrow) after cells had migrated into the void. Migration distances were measured by drawing a line from the edge of the scrape at time 0 to the leading edge of the most migratory cells after 24h. B ) Schematic representation of transwell motility chambers. Cells are plated onto the upper membrane and migrate through the porous membrane to the lower chamber. C) Schematic representation of transwell invasion chambers. Transwell membranes were coated with 5% gelatin prior to experiments. Cells are plated onto the upper membrane and invade through the gelatin and porous membrane to the lower chamber.  A  44  Figure 3.6 Analysis of C 6 subclone motility by wound healing. A) Micrographs o f subclones 24h following wounding. The red dashed line indicates the site o f the lesion. B ) Measurements of migration distances indicated that C 6 - H cells had moved a distance significantly greater than C 6 - L subclones 24h following wounding (n=3, P<0.01).  46  Figure 3.7 Analysis of C6 subclone motility by transwell. The percentage o f C 6 - H cells that migrated to the underside o f uncoated P E T membranes was significantly greater than that o f C 6 L subclones afterHh (n=3, P<0.0001).  48  Figure 3.8 Efficiency of endogenous Cx43 knockdown in C6 cells. A ) Cx43 protein expression after infection with Cx43 s h R N A - 1 (but not C x 4 3 s h R N A - 2 ) was reduced compared to both empty and scrambled controls as detected by Cx43 immunoreactivity of whole cell lysates subjected to western blot analysis. B ) Densitometric analysis indicated that the intensity o f Cx43 expression in C6-Cx43 shRNA-1 cells was reduced by approximately 3 fold (36%) compared to cells expressing scrambledCx43shRNA following knockdown (n=2).  #  #  „ f  of  y  kDa  43 37  B  Cx43 GAPDH  Figure 3.9 Knockdown of endogenous Cx43 in C6 cells  Consistent with Western immunoblotting, immunocytochemistry assays indicate that scrambled control cells express higher levels of Cx43 than C x 4 3 s h R N A cells. Note that exposure times for both scrambled and Cx43 shRNA cells labeled with antibodies against  Cx43 were the  same.  52 Figure 3.10 Knockdown of Cx43 by shRNA decreases directional motility. In wound healing assays cells stably expressing Cx43 s h R N A - 1 moved a distance significantly less than empty or scrambled control 24h following wounding (n=9, P<0.05). Note: N o significant differences were observed between empty, scrambled, C x 4 3 s h R N A - 2 , and untransfected parental controls (P>0.05)  empty  scrambled  Cx43shRNA-1  Cx43shRNA-2  54  Figure 3.11 Knockdown of Cx43 by shRNA decreases non-directional motility. In transwell motility assays fewer Cx43 shRNA-1 infected cells migrated through uncoated transwell membranes (8 u m pore size) than scrambled control after lOh (n=3, representative image shown)  56 F i g u r e 3.12 K n o c k d o w n of C x 4 3 by s h R N A decreases invasion. In transwell invasion assays fewer Cx43 s h R N A infected cells invaded through gelatin coated transwell membranes (8 um pore size) than scrambled control after lOh (n=3, representative image shown)  58  Figure 3.13 Knockdown of Cx43 by shRNA reduces gap junction intercellular communication. Gap junction intercellular communication was assessed in confluent cultures by the preloading method (see text for details). Dye transfer was restricted in cells expressing C x 4 3 s h R N A - l compared to cells expressing scrambled control.  59  Scrambled DIC/FITC/Rhodamine overlay  Cx43shRNA FITC/Rhodamine overlay  60 Figure 3.14 Efficacy of the gap junction blocker carbenoxolone. Preloading studies indicate that the passage o f calcein dye was prevented in cultures incubated with 1 5 0 u M carbenoxolone (top). The same concentration o f the inactive analogue glyccccherrizzhicctic acid did not impede dye passage (bottom). Bar =50um.  CBX- Gap Junction Blocker DIC/FITC/Rhodamine overlay  FITC/Rhodamine overlay  GZA- Inactive Analog DIC/FITC/Rhodamine overlay  FITC/Rhodamine overlay  62 Figure 3.15 Blocking gap junction intercellular communication did not alter directional motility. Wound healing assays were conducted on C 6 - H cells in the presence o f 150 u M C B X . The migration distance 24h after wounding was not altered compared to untreated or 150 u M G Z A controls (n=6, P>0.05).  63  800 -j 3.  700 600 -  o E o i_ o E o> u c (0 w  -a  500 400 300 200 100 -  0Untreated  GZA  CBX  64 Figure 3.16 Blocking gap junction intercellular communication did not alter nondirectional motility. Transwell motility assays were conducted on C 6 - H cells in the presence of 150 u M C B X . There was no significant difference in the migration o f cells treated with gap junction blocker compared to untreated or 150 u M G Z A controls after incubation for 14h (n=3, P>0.05).  migration index (normalized to C 6 - h r o - ^ 0 5 0 o o r o - u o > o o o o o o o o  cn  66 Figure 3.17 Levels of exogenous Cx43 are similar between cells expressing truncated and full length forms of Cx43. Whole cell lysates from C 6 cells were run on polyacrylamide gels and probed for A ) G F P to indicate exogenous G F P tagged C x 4 3 or B ) C x 4 3 to indicate endogenous C x 4 3 . G A P D H served as a loading control.  Cx43GFP  kDa 70 50  37  Cx43GFP Cx43A244-382GFP  GAPDH  aGFP  68  Figure 3.18 Localization of truncated Cx43 in C6 cells. A ) In live cultures, gap junction plaques composed o f truncated Cx43 tagged with G F P were apparent at the plasma membranes between adjacent cells. B ) In fixed cultures, endogenous Cx43 was detected using a mouse polyclonal antibody directed against the carboxy terminus o f Cx43 (Red). Truncated Cx43 ( G F P , green) localized to the outer membrane and formed homomeric plaques (green) or heteromeric plaques with endogenous Cx43 (yellow).  Figure 3.19 C6 cells expressing truncated Cx43 are communication competent. Dye preloading indicated that C6-Cx43A244-382GFP cells exhibit gap junction intercellular communication.  71  72  Figure 3.20 Enhanced directional motility requires full lenghth Cx43. Exogenous expression of full length Cx43 significantly enhanced directional motility compared to control (p<0.05). Expression of Cx43A244-382GFP, however, did not exhibit altered motility compared to control cells (P>0.05).  74 Figure 3.21 Truncation of the Cx43 carboxy terminus does not alter non-directional motility. Exogenous expression of full length C x 4 3 G F P slightly but not significantly enhanced non-directional cell motility compared to control. Similarly, the expression o f Cx43A244382GFP did not result in altered motility compared to control cells (P>0.05, n=5).  76 Figure 3.22 Expression of Cx43A244-382GFP results in an altered motility phenotype. Micrographs of cells imaged 24h after wounding indicate that while control cells (A) and cells expressing full length Cx43 (B) migrated away from the site o f the lesion as individual cells (arrows), cells expressing truncated Cx43 (C) migrated away from the wound site as a collective sheet.  C6-GFP  C6-Cx43GFP  C6-Cx43ACTGFP  78 Figure 3.23 Expression and localization of N-Cadherin is not altered in C6 cells expressing Cx43ACT-GFP. A ) The abundance o f N - C a d protein is not altered in between C6-Cx43A244382GFP and control cells. B) N-Cadherin is localized to the periphery o f the plasma membrane at sites o f cell-cell contact in both control (upper) and C6-Cx43A244-382GFP expressing (lower) cells (arrows).  A  kDa  o  o  v N-Cad  120  37 GAPDH  B  C6-Cx43GFP  C6-Cx43ACTGFP  80 Figure 3.24 Expression and localization of pl20ctn is not altered in C6 cells expressing Cx43ACT-GFP. A ) The abundance o f pl20ctn protein is not altered between C6-Cx43A244382GFP and control cells. B ) p l 2 0 c t n is localized in the cytoplasm, nucleus, and to the periphery of the plasma membrane at sites o f cell-cell contact in both control (upper) and C6-Cx43A244382GFP expressing (lower) cells (arrows).  81  Cx43ACT-GFP  82 C H A P T E R 4 DISCUSSION 4.1 S U M M A R Y The rationale for studying Cx43 in the context o f glioma migration is two fold. First, it has been demonstrated that Cx43 expression is reduced in high grade glioma (Pu et al., 2004), (Soroceanu et al., 2001; Huang et al., 1999). Second, G J I C may be responsible for the observation that the invasion o f malignant glioma into brain parenchyma concomitantly induces the phenotypic transformation o f astrocytes, as evidenced by reactive gliosis in astrocytes surrounding the invasive tumor cells (Knott et al., 1998; L e et al., 2003). It is clear that the brain tissue plays a critical role, whether complaisant or resistant, for glioma invasion. Since Cx43 is the most prevalent C x in astrocytes, gap junction channels composed o f Cx43 likely constitute a major contribution to tumor-astrocyte interactions. Indeed direct intercellular communication between C 6 glioma and astrocytes in vitro, as well as astrocytes in vivo, was increased when C6 cells were engineered to overexpress Cx43 (Zhang et al 1999, 2003, L i n et al 2002). To date, only 3 studies have directly examined the role o f Cx43 in glioma motility and/or invasion ( L i n et al., 2003b; Oliveira et al., 2005; Zhang et al., 2003a). These studies have employed an exogenous overexpression approach which impose confounding variables such as defects in Cx43 assembly and trafficking (VanSlyke, 2005). To circumvent these variables, we took a novel approach and compared the motility o f C 6 subclones expressing high levels o f endogenous Cx43 to parental C 6 subclones expressing low levels o f C x 4 3 . T o ensure that additional differences between these subclones (i.e. other than Cx43 expression) did not account for differences observed in functional motility assays, Cx43 expression, as determined by western blot and immunocytochemistry, was suppressed using s i R N A . The results o f these experiments indicated that Cx43 enhanced motility and invasion in C 6 cells which is in agreement with other studies ( L i n et al., 2002; Oliveira et al., 2005; Zhang et al., 2003a) that investigated the role o f Cx43-mediated motility and invasion in these cells.  83 In addition to a reduction o f Cx43 protein upon s h R N A knockdown, the preloading method o f dye transfer (Goldberg et al., 1995) indicated that gap junction intercellular coupling is impaired in cells expressing C x 4 3 s h R N A - l compared to scrambled control. It was hence reasoned that the decreased cell motility observed could be a consequence o f decreased G J I C or Cx43 expression independent o f GJIC. T o clarify these confounding factors, the gap junction blocker carbenoxolone was added to the culture medium during wound healing and transwell motility assays. N o significant differences in directional or random motility could be detected between C B X , inactive control G Z A , or untreated cells. W h i l e C B X is a non-specific blocker o f gap junctions (Blomstrand et al., 2004), persistence o f a function (i.e. motility) in the presence o f this blocker is quite good evidence that G J I C does not contribute significantly to motility in C 6 cells (Bennett and Zukin, 2004). Therefore, it is possible that Cx43-mediated motility may be attributed to actions o f Cx43 that do not involve GJIC. The Cx43 C-terminus is thought to mediate the interaction with the majority o f C x 4 3 interacting proteins (Giepmans, 2004; Herve et al., 2004a). T o consider the possibility that protein interactions with the carboxy terminus o f Cx43 mediate the role o f Cx43 in cell motility, a comparison o f both directional and random motility was undertaken in communication competent C 6 cells exogenously expressing full length or truncated C x 4 3 . These cells were generated by the retroviral infection o f construncts into C 6 cells expressing low levels o f Cx43 (however not the C 6 - L cells described earlier) and were not subjected to subsequent selection (see F u et al., 2004). Random motility, evaluated by transwell chambers, was not significantly different between these cells. In wound healing assays cells expressing full length Cx43 were significantly more motile than both C 6 - G F P control and C 6 - C x 4 3 A C T 2 4 4 - 3 8 2 G F P cells, while there was no significant difference in the wound healing ability o f cells expressing truncated Cx43 compared to control. This indicates that the Cx43 C-terminus likely plays an important role in the ability o f Cx43 to enhance motility by a mechanism that does not impair full length C x 4 3 .  84 The collective sheet-like migration exhibited by C 6 - C x 4 3 A C T 2 4 4 - 3 8 2 G F P cells suggested that the loss o f the Cx43 C-terminus may correlate with increased adhesion. In an effort to elucidate differences between C 6 - C x 4 3 G F P and C 6 - C x 4 3 A C T 2 4 4 - 3 8 2 G F P cells with respect to adhesion, N - C a d and pl20ctn, Cx43 associating proteins with known roles in motility and adhesion, were assessed for protein expression and localization by western blot and immunocytochemistry respectively. W e were unable to detect any changes in total protein expression o f either N - C a d or pl20ctn between C 6 - C x 4 3 G F P and C 6 - C x 4 3 A C T 2 4 4 - 3 8 2 G F P cells. Similarly, in both cell types these proteins localized to the cell membrane, nucleus, and cytoplasmic component. 4.2 C x 4 3 - M E D I A T E D I N V A S I O N I N H O M O C E L L U L A R P O P U L A T I O N S O F C6 C E L L S  Several studies have attempted to elucidate the role o f Cx43 in glioma motility and/or invasion. Zhang et al. (2003) observed increased chemokinesis o f C 6 cells overexpressing Cx43 (C6-Cx43) compared to mock transfected control cells (C6-mock) in radial dish assays (Zhang et al., 2003a). Using transwell chambers to evaluate chemokinesis, we noted a decrease in cell motility upon Cx43 knockdown with s h R N A . Together, these studies suggest that Cx43 expression correlates positively with chemokinesis. Zhang et al (2003) further compared the invasivity o f C 6 - C x 4 3 versus C6-mock cells in the presence or absence o f astrocytes. In the absence o f astrocytes, the rate o f invasion was similar between C 6 - C x 4 3 and C6-mock cells while C 6 - C x 4 3 cells were much more invasive in the co-culture experiments. Conversely, in the present study we found that Cx43 knockdown dramatically attenuated invasion in homocellular populations o f C 6 cells, suggesting that the modulation o f Cx43 levels alone are able to alter invasivity independent o f heterocellular interactions. Although both studies employed the in vitro transwell invasion system, there are two notable differences between the assays. First, Zhang et al. (2003) employed C 6 cells overexpressing Cx43 while we used C 6 cells in which high endogenous levels o f Cx43 were knocked down using s h R N A . Both o f these  85 approaches may introduce confounding variables as hemichannels are assembled prematurely in the E R o f C 6 cells overexpressing Cx43 (VanSlyke, 2005) and s h R N A may have off target effects. Additionally, the C 6 subclones employed in these studies may not have been derived from the same parental cells. Given the heterogeneity o f the C 6 line, it is possible that differences unique to the cells employed in the respective studies gave rise to these differences. Secondly, despite having demonstrated that a consequence o f Cx43 overexpression was an increase in the level o f gelatinase activity (Zhang et al 2003), Zhang et al (2003) used Matrigel as an invasive substrate. Matrigel, which is artificially reconstituted basement membrane, contains numerous and undefined growth factors and signaling molecules which confound in vitro analysis (Vukicevic et al., 1992). W e employed gelatin as an invasive substrate which is devoid o f such undefined growth factors and signaling molecules and is the E C M component that is preferentially degraded by the gelatinases. The former study therefore examines chemotactic properties while the latter examines chemokinetic properties. Notably, neither o f these invasion experiments was conducted in the presence o f gap junction blockers; therefore, the contribution o f G J I C was not assessed. Together these studies suggest that while the co-culture o f C 6 glioma cells and astrocytes may indeed increase the invasivity o f tumor cells exogenously overexpressing C x 4 3 , co-culture may not be essential for C 6 invasion since knockdown o f Cx43 expression reduces the invasivity of homocellular populations o f C 6 cells. These studies additionally caution that because the invasive substrate is an important determinant o f cellular invasion, equivalent substrates must be employed when comparing results between studies to avoid confounding factors. Furthermore, both studies support a motility and invasion enhancing role for Cx43 in C 6 cells. The contribution o f Cx43 in heterocellular interactions between C 6 glioma and astrocytes was investigated in vivo by L i n et al. (2002). C6-Cx43 cells implanted into the striatum o f Wistar rats exhibited widespread passage o f Lucifer yellow to host astrocytes. In contrast, C6-mock and  86 C 6 cells expressing chimeric Cx40/Cx43 (Cx40*43C3) proteins created b y splicing the C terminus o f Cx43 into Cx40 (Haubrich et al., 1996), failed to establish Lucifer yellow coupling with surrounding host cells. Coincident with decreased G J I C , both C6-mock and C6-Cx40*43C3 cells exhibited adluminal invasion while C6-Cx43 cells aggressively infiltrated the brain parenchyma. The use o f the communication-incompetent Cx40*43C3 protein in these studies suggests a requirement for G J I C in glioma invasion. 4.3 Cx43 A N D G J I C I N G L I O M A C H E M O K I N E S I S A N D C H E M O T A X I S The contribution o f G J I C in heterocellular glioma motility was evaluated using established human glioblastoma cell lines in addition to C 6 cells in several in vitro assays (Oliveira, 2005). In an ex vivo brain slice assay heterocellular G J I C between brain slices and C 6 cells could be manipulated throughout chemokinesis using pharmacological blockers. Blocking GJIC by C B X in the cell type that exhibited the most extensive heterocellular coupling ( G L 1 5 cells) decreased the migration distance as well as the number o f migrating cells compared to G Z A treated controls (Oliveira, 2005). B y contrast, in the current study blocking Cx43 channels with C B X did not alter chemotaxis or chemokinesis in homocellular populations o f C 6 cells. Wound healing assays are a particularly relevant in vitro method to study the motility o f glioma cells as lesions are inflicted, as they are in surgical resection. The lesion provides uniform directionality to motile cells. The imposed insult when lesioning confluent monolayers o f cells provides a chemorepulsive stimulus, directing cells away from the site o f injury (Sato and Rifkin, 1988). Although it is difficult to detect a chemorepulsive or chemoattractive gradient directly, it has been shown that cells hundreds o f microns from the wound edge extend lamellipodia beneath cells in front o f them toward the lesion. (Farooqui and Fenteany, 2005). Importantly, the authors found that this effect persisted in the presence o f gap junction blockers. Furthermore, increased production o f motility enhancing E C M proteins such as tenascin and thrombospondin occurs upon injury to the epithelium (Majesky, 1994). These stimuli down-regulate the assembly and  87 activity o f focal adhesions (Zagzag et al., 2002) and may well override motility-altering effects caused by blocking GJIC. This may explain why no significant differences i n wound healing ability were detected in C B X treated cultures compared to G Z A treated or untreated control. The role o f G J I C in chemotaxis might then be better studied using the transwell assay. However, in the transwell assay it is not clear whether cells indeed form gap junctions with adjacent cells in the absence o f gap junction uncouplers and therefore blocker treatment may not have functional consequences on GJIC-mediated motility. B y implanting a mixed population o f C 6 and C 6 C x 4 3 G F P cells into rat brain, Peschanski et al (2005) found that C 6 - C x 4 3 G F P cells were preferentially located at the periphery o f the tumor mass while C 6 cells expressing low levels o f Cx43 were present only in the tumor mass. This finding is consistent with the previous studies mentioned that examined the role o f Cx43 in glioma motility, as well as by the current study in which C 6 - C x 4 3 cells exhibited increased chemotaxis compared to control (C6-GFP) cells. In summary, the studies that have directly examined the role o f Cx43 in glioma motility and/or invasion demonstrate a positive correlation between these cellular processes and Cx43 expression, although the contribution o f G J I C has not yet been conclusively elucidated. 4.4 I N V O L V E M E N T O F G J I C  Although few studies conducted thus far have directly examined the role o f Cx43 in glioma motility, a number o f studies have directly investigated the role o f Cx43 in other cell types. M a n y o f these studies indicate a positive correlation between Cx43 and motility and/or invasion. The role o f G J I C in transendothelial migration (diapedesis) o f mammary epithelial tumors was investigated using a C x - and GJIC-deficient, non-metastatic cell line ( H B L 1 0 0 ) in conjunction with Cx43-expressing human microvascular endothelial cells ( H M V E C s ) derived from the lung (Pollmann et al., 2005). H B L 1 0 0 cells engineered to express wild-type Cx43 exhibited enhanced diapedesis while cells expressing a non-functional (i.e. G J I C incompetent) chimeric mutant o f Cx43 were similar to control (Pollmann et al., 2005). G J I C was tested  88 directly b y blocking both homocellular and heterocellular G J I C with C B X i n co-cultures and resulted in reduced diapedesis o f Cx43 expressing H B L 1 0 0 tumor cells (Pollmann et al., 2005). A s these results indicate a positive correlation between Cx43 and invasion, they support the results found in studies investigating Cx43 in glioma invasion, including the results obtained in the present study, and further suggest the involvement o f heterologous G J I C in cellular invasion. The regulation o f heterocellular G J I C may be more important than a straightforward communication-competent versus communication-incompetent scenario. The deletion o f the Cx43 gene, Gjal, in mice results in perinatal lethality due to conotruncal heart malformations and pulmonary outflow obstruction (Reaume et al., 1995). These regions o f the heart are populated by migratory cardiac neural crest cells emanating from the dorsolateral margins o f the neural tube. In an ex vivo culture system, explants o f the neural tube from transgenic mice ( C x 4 3 K O or C M V 4 3 , a transgenic mouse line in which Cx43 is overexpressed) revealed that changes in the level o f G J I C correlated positively with parallel changes in the rate o f neural crest migration (Huang et al., 1998a). Furthermore, blocking G J I C with oleamide significantly reduced outgrowth (Huang et al., 1998a). In vivo the hearts o f C M V 4 3 mice revealed an increased abundance o f neural crest cells in the outflow septum while the hearts o f C x 4 3 K O mice exhibited an obvious thinning o f the conotruncal myocardium and the presence o f fewer neural crest cells (Huang et al., 1998a). Importantly, the hearts o f C M V 4 3 mice also suffered from malformations o f the conus region and outflow tract obstructions (Ewart et al., 1997; Huang et al., 1998b) despite exhibiting high levels o f G J I C (Huang et al., 1998a). These studies suggest that while the migration o f neural crest cells indeed requires G J I C , the precise regulation of G J I C may be o f critical importance for conotruncal heart development. Such precise regulation is intimately related to the Cx43 C-terminus. Specifically, Xenopus laevis oocytes expressing C-terminal truncated Cx43 were resistant to G J I C uncoupling agents such as insulin and insulin-like growth factor (IGF; Homma et al., 1998) and v-src (Zhou et al., 1999). In murine  89 neuroblastoma (N2a) cells, single channel analysis indicated that truncation o f the Cx43 C terminus (Cx43M257 mutant) domain did not significantly modify the magnitude o f the main unitary conductance o f Cx43 channels, while the mean open time o f C x 4 3 M 2 5 7 channels was considerably prolonged compared to full length channels (2450±200 ms compared to 126±20 ms respectively; Moreno et al., 2002). It is therefore important to consider that G J I C may account for differences when assessing cellular functions in cells expressing mutated and/or truncated Cx43 C-termini. In our study, the motility o f C 6 cells exogenously expressing Cx43 in which the C-terminus had been deleted (C6-Cx43A244-382GFP) was not significantly altered compared to control cells, however the exogenous expression o f full length Cx43 significantly enhanced directional motility compared to control and C6-Cx43A244-382GFP cells which suggests that the Cx43 C-terminus plays an important role in enhancing directional motility in homocellular cell populations in vitro. A s the C6-Cx43A244-382GFP cells exhibited dye coupling, it is possible that the observed disturbances in differences in directional motility may be attributed to GJIC. The use o f blockers throughout the motility experiments using C6-Cx43A244-382GFP cells is essential to rule out the possibility that the Cx43 C-terminus is involved in glioma invasion independently o f gap junction channel formation. 4.5 I N V O L V E M E N T O F T H E C x 4 3 C - T E R M I N U S IN C H E M O K I N E S I S Moorby et al. (2000) recognized that low-molecular weight molecules permeable to gap junctions may account for the reduction in motility they observed i n mouse 3T3 A 3 1 fibroblasts expressing a communication-competent, truncated Cx43 mutant (Cx43-256M, described in (Moorby and Gherardi, 1999) compared to w i l d type 3T3 A 3 1 cells which normally express abundant Cx43 and have a high basal level o f G J I C (Moorby, 2000). The expression o f C x 4 3 2 5 6 M prevented PDGF-induced inhibition o f G J I C (Moorby and Gherardi, 1999) and also prevented enhanced chemokinesis in response to P D G F (Moorby, 2000). Control cells, on the other hand, exhibited a significant increase in chemokinesis upon P D G F stimulation and  90 resultant impairment o f GJIC. Conversely we did not see changes in chemokinesis upon blocking G J I C with C B X . W h i l e C B X is a non-specific gap junction inhibitor, its inactive analog, G Z A , offers a control in addition to the untreated condition. In contrast P D G F has an overwhelming number o f effects in addition to blocking G J I C which may confound the interpretation o f abolishing GJIC. Chemokinesis was not altered, however, in cells expressing C x 4 3 - 2 5 6 M compared to wild-type 3T3 A 3 1 cells which is consistent with our finding that C 6 cells expressing Cx43A244-382GFP did not exhibit altered chemokinesis compared to C 6 - C x 4 3 G F P cells. The C-terminus truncation o f Cx43 alters G J I C yet neither our study nor M o o r b y et al. (2000) were able to detect alterations in chemokinesis as a result o f this truncation. Furthermore, as our results, which directly examine the consequence o f blocking C x 4 3 channels, did not indicate a change in chemokinesis, it is likely that G J I C is not involved i n regulating chemokinesis o f C 6 cells. W h i l e the results o f L i n et al. suggest that the Cx43 C-terminus i n the absence o f G J I C does not enable parenchymal invasion, the chimeric Cx40*43C3 protein employed in their studies may not accurately mimic the function o f endogenous Cx43 protein sufficiently for the proper deployment o f downstream Cx43 signaling cascades which may be independent o f Cx43 gap junction formation. Although it is true that the majority o f Cx43 interacting proteins for which Cx43 binding sites are known to associate with the C-terminus o f C x 4 3 , the specific sites o f interaction for many associating proteins remains to be determined. Similarly, the interaction sites o f other Cx43-associating proteins are not known. Furthermore, the functions o f the Cx43 C-terminus may depend on interactions with the Cx43 intracellular loop domain (Seki et al., 2004). The use o f the non-conducting Cx43 mutant which possesses an intact C-terminus ( L i n et al., 2003b) would likely better elucidate the contribution o f G J I C to Cx43-mediated mechanisms of glioma invasion in these studies.  91 4.6 T H E I N V O L V E M E N T O F T H E Cx43 C - T E R M I N U S IN C H E M O T A X I S In response to a lesion stimulus in wound healing assays, M o o r b y et al. (2000) observed decreased chemotaxis in cells expressing truncated C x 4 3 . Similarly, we found a significant difference in chemotaxis in cells expressing full length Cx43 compared to cells expressing truncated C x 4 3 . However, we did not observe a significant difference in chemotaxis between C 6 G F P and C6-Cx43A244-382GFP cells. Only Moorby et al. (2000) and the current study have directly examined the consequence o f Cx43 C-terminus truncation on motility, albeit using different constructs in different cells (Cx43-256M and Cx43A244-382GFP in 3T3 A 3 1 and C 6 cells, respectively). This present work does not aim to uncover the molecular mechanisms by which chemotaxis and chemokinesis are distinct from one another, but rather recognizes that the parameters employed in this study to investigate motility are not equal in that they measure different aspects o f cell motility which have been classified as chemotaxic or chemokinetic. However, because neither study was able to detect significant differences i n the random motility of cells expressing full length Cx43 or truncated C x 4 3 , it suggests that cells expressing Cx43 which lacks the C-terminus may not express or may not be able to recruit the necessary cellular machinery required for directional motility imposed by lesion stimulus. Indeed C6-Cx43A244382GFP cells migrated away from the lesion as a collective sheet unlike C 6 - C x 4 3 G F P cells which did so as detached, individual cells. Although we found no differences in the total expression or sub-cellular localization o f either N - C a d or pl20ctn, analyzing these parameters under lesioned conditions may yield different results. Additional support for the hypothesis that the Cx43 C-terminus is involved in chemotactic response comes from the comparison o f C 6 Cx43shRNAscrambled and C 6 - C x 4 3 s h R N A - 1 cells i n the present study. Significant differences existed between these cells for both chemotaxis and chemokinesis. A s differences in random motility were greater than differences in directional motility, it suggests that the response to the lesion stimulus may be similar in these cells, perhaps owing to the intact Cx43 C-terminus in  92 both cell types. Indeed Cx43 has been shown to regulate polarized cell movement essential for directional migration in cardiac neural crest cells ( X u et al., 2006). Whether this proposed scenario would be mediated via a gap junction dependent or independent mechanism can not be ascertained from the experiments conducted in our study or those o f M o o r b y et al. (2002).  4.7 T H E R O L E O F Cx43 IN M O T I L I T Y C O U L D B E C E L L - T Y P E SPECIFIC Although the majority o f studies investigating the role o f Cx43 in cell motility and/or invasion have found that Cx43 correlates positively with these processes, several studies have found an inverse correlation. Using MD-831 breast carcinoma cells in transwell migration chambers, Shao et al. (2005) observed reduced chemokinesis upon Cx43 knockdown by s h R N A . Q i u et al (2003) found increased rate o f wound closure upon Cx43 knockdown by s i R N A while Brandner et al. (2004) noted persistence o f Cx43 at the margins o f non-healing wounds and a loss o f Cx43 staining at wound margins during initial stages o f wound healing (Brandner et al., 2004). Indeed wound closure in C x 4 3 K O mice was realized more quickly compared to wild-type controls (Kretz et al., 2003). These studies compliment one another and together indicate that the role o f Cx43 in cell motility may be cell-type specific. 4.8 C O N C L U S I O N S A N D P H Y S I O L O G I C A L R E L E V A N C E In conclusion, the majority o f studies indicate that Cx43 expression correlates positively with chemotaxis. In vitro assays indicate a significant role for the Cx43 C-terminus in this process; however the mechanism by which it might be achieved has not been investigated. The use o f gap junction blockers has not yet provided a clear indication o f whether GJIC is involved in Cx43-mediated chemotaxis. N o consensus has been reached for the role or mechanism o f Cx43 with respect to chemokinesis, although the study by Shao et al. (2005) is the only study to date that has indicated altered chemokinesis as a result o f Cx43 expression. W h i l e heterocellular coupling may facilitate glioma invasion, decreased Cx43 expression attenuates invasion in homocellular populations and therefore heterocellular coupling may not be a requirement for  93 Cx43-mediated invasion. This finding is particularly relevant as it implies that cells in the tumor core, a homocellular environment, may exhibit increased exodus concomitant with increased expression o f C x 4 3 . A s numerous growth factors and cytokines to which cells in the tumor core may be exposed have been reported to upregulate Cx43 (Ozog et al., 2002), such an implication offers to reconcile the apparent paradox that while Cx43 expression is downregulated in high grade gliomas which are notoriously invasive, reconstitution o f Cx43 to levels typical o f normal astrocytes enhances motility.  94 4.9 F U T U R E D I R E C T I O N S 4.9.1  Non-conducting Cx43 mutants could be used to evaluate the role o f G J I C in glioma motility and/or invasion. These experiments could be performed using transwell chambers in the presence o f a chemotaxic stimulus to differentiate between chemokinesis and chemotaxis, or alternatively in ex vivo assays (below).  4.9.2  The tumor environment may well contribute to tumor phenotype. The use o f ex vivo brain slice invasion assays from Cx43 " mice could be employed to evaluate _/  the role o f C x 4 3 , and possibly G J I C , in the stromal environment during glioma invasion. 4.9.3  C6-Astrocyte co-culture experiments could be undertaken by plating both C 6 cells and astrocytes in the same culture dish. A wound inflicted such that the migrating cells collide with one another would allow analysis o f whether the Cx43 levels in the C 6 cells change upon forming heterocellular contacts. (Note that C 6 cells can be differentiated from astrocytes as C 6 cells stain positively for vimentin while astrocytes do not).  4.9.4  The use o f Cx43 C-terminal mutants to explore the interaction o f Cx43 with Cx43 associating proteins may indicate which Cx43-associating proteins are involved in Cx43-mediated motility.  4.9.5  Protein analysis (ICC, Western, Protein Array) o f cells at the wounded edge compared to cells in the confluent sub-marginal marginal zone.  95 REFERENCES Aberle, H . , H . Schwartz, and R. Kemler. 1996. Cadherin-catenin complex: protein interactions and their implications for cadherin function. J Cell Biochem. 61:514-23. Ackert, C . L . , J.E. Gittens, M . J . O'Brien, J.J. Eppig, and G . M . Kidder. 2001. Intercellular communication via connexin43 gap junctions is required for ovarian folliculogenesis in the mouse. Dev Biol. 233:258-70. A i , Z . , A . Fischer, D . C Spray, A . M . Brown, and G.I. Fishman. 2000. Wnt-1 regulation o f connexin43 i n cardiac myocytes. / Clin Invest. 105:161-71. Anand-Apte, B . , and B . Zetter. 1997. Signaling mechanisms i n growth factor-stimulated cell motility. Stem Cells. 15:259-67. Anastasiadis, P.Z., S . Y . M o o n , M . A . Thoreson, D . J . Mariner, H . C . Crawford, Y . Zheng, and A . B . Reynolds. 2000. Inhibition o f R h o A b y p l 2 0 catenin. Nat Cell Biol. 2:637-44. Anastasiadis, P.Z., and A . B . Reynolds. 2001. Regulation o f Rho GTPases b y pl20-catenin. Curr Opin Cell Biol. 13:604-10. Asada, H . , K . Uyemura, and T. Shirao. 1994. Actin-binding protein, drebrin, accumulates i n submembranous regions in parallel with neuronal differentiation. J Neurosci Res. 38:14959. Avanzo, J.L., M . Mesnil, F.J. Hernandez-Blazquez, Mackowiak, II, C M . M o r i , T C . da Silva, S C . Oloris, A . P . Garate, S . M . Massironi, H . Yamasaki, and M . L . Dagli. 2004. Increased susceptibility to urethane-induced lung tumors i n mice with decreased expression o f connexin43. Carcinogenesis. 25:1973-82. Azarnia, R., M . Mitcho, D . Shalloway, and W . R . Loewenstein. 1989. Junctional intercellular communication is cooperatively inhibited b y oncogenes i n transformation. Oncogene. 4:1161-8. Bani-Yaghoub, M . , J . M . Felker, C . Sans, and C C Naus. 2000. The effects o f bone morphogenetic protein 2 and 4 ( B M P 2 and B M P 4 ) on gap junctions during neurodevelopment. Exp Neurol. 162:13-26. Bani-Yaghoub, M . , T . M . Underhill, and C C . Naus. 1999. Gap junction blockage interferes with neuronal and astroglial differentiation o f mouse P19 embryonal carcinoma cells. Dev Genet. 24:69-81. Batias, C , J.P. Siffroi, P. Fenichel, G . Pointis, and D . Segretain. 2000. Connexin43 gene expression and regulation in the rodent seminiferous epithelium. J Histochem Cytochem. 48:793-805. Beardslee, M . A . , J . G . Laing, E . C . Beyer, and J.E. Saffitz. 1998. Rapid turnover o f connexin43 in the adult rat heart. Circ Res. 83:629-35. Bellovin, D.I., R . C . Bates, A . Muzikansky, D . L . R i m m , and A . M . Mercurio. 2005. Altered localization o f p i 20 catenin during epithelial to mesenchymal transition o f colon carcinoma is prognostic for aggressive disease. Cancer Res. 65:10938-45. Benda, P., J. Lightbody, G . Sato, L . Levine, and W . Sweet. 1968. Differentiated rat glial cell strain in tissue culture. Science. 161:370-1. Benini, S., B . Perbal, D . Zambelli, M . P . Colombo, M . C . Manara, M . Serra, M . Parenza, V . Martinez, P. Picci, and K . Scotlandi. 2005. In Ewing's sarcoma C C N 3 ( N O V ) inhibits proliferation while promoting migration and invasion o f the same cell type. Oncogene. 24:4349-61. Benjamin, R., J . Capparella, and A . Brown. 2003. Classification o f glioblastoma multiforme in adults b y molecular genetics. Cancer J. 9:82-90. Bennett, M . V . 1997. Gap junctions as electrical synapses. JNeurocytol. 26:349-66.  96 Bennett, M . V . , and R.S. Zukin. 2004. Electrical coupling and neuronal synchronization in the Mammalian brain. Neuron. 41:495-511. Bergoffen, J., S.S. Scherer, S.Wang, M . O . Scott, L . J . Bone, D . L . Paul, K . C h e n , M . W . Lensch, P.F. Chance, K . F . Fishbeck. 1993. Connexin mutations i n X - l i n k e d Charcot-Marie-Tooth disease. Science. 262(5142):2039-42 Berthoud, V . M . , P.J. Minogue, J . G . Laing, and E . C . Beyer. 2004. Pathways for degradation o f connexins and gap junctions. Cardiovasc Res. 62:256-67. Bertram, J.S., and A . L . Vine. 2005. Cancer prevention by retinoids and carotenoids: independent action on a common target. Biochim Biophys Acta. 1740:170-8. Beyer, E . C , D . L . Paul, and D . A . Goodenough. 1990. Connexin family o f gap junction proteins. JMembrBiol. 116:187-94. Blomstrand, F., L . Venance, A . L . Siren, P. Ezan, E . Hanse, J. G l o w i n s k i , H . Ehrenreich, and C . Giaume. 2004. Endothelins regulate astrocyte gap junctions in rat hippocampal slices. Eur J Neurosci. 19:1005-15. Brandner, J . M . , P. Houdek, B . Husing, C . Kaiser, and I. M o l l . 2004. Connexins 26, 30, and 43: differences among spontaneous, chronic, and accelerated human wound healing. J Invest Dermatol. 122:1310-20. Brindle, N . P . , M . R . Holt, J.E. Davies, C.J. Price, and D . R . Critchley. 1996. The focal-adhesion vasodilator-stimulated phosphoprotein ( V A S P ) binds to the proline-rich domain in vinculin. Biochem J. 318 ( Pt 3):753-7. Brink, P.R., K . Cronin, K . Banach, E . Peterson, E . M . Westphale, K . H . Seul, S . V . Ramanan, and E . C . Beyer. 1997. Evidence for heteromeric gap junction channels formed from rat connexin43 and human connexin37. Am J Physiol. 273:C1386-96. Bruzzone, R., T . W . White, and D . L . Paul. 1996. Connections with connexins: the molecular basis o f direct intercellular signaling. Eur J Biochem. 238:1-27. Bukauskas, F.F., K . Jordan, A . Bukauskiene, M . V . Bennett, P . D . Lampe, D . W . Laird, and V . K . Verselis. 2000. Clustering o f connexin 43-enhanced green fluorescent protein gap junction channels and functional coupling in living cells. Proc Natl Acad Sci USA. 97:2556-61. Butkevich, E . , S. Hulsmann, D . Wenzel, T. Shirao, R. Duden, and I. Majoul. 2004. Drebrin is a novel connexin-43 binding partner that links gap junctions to the submembrane cytoskeleton. Curr Biol. 14:650-8. C B T R U S . http://www.cbtrus.org/reports/reports.html Chatterjee, B . , R . A . Meyer, G . A . Loredo, C M . Coleman, R . Tuan, and C W . L o . 2003. B M P regulation o f the mouse connexin43 promoter in osteoblastic cells and embryos. Cell Commun Adhes. 10:37-50. Chen, Z . Q . , D . Lefebvre, X . H . B a i , A . Reaume, J. Rossant, and S.J. L y e . 1995. Identification o f two regulatory elements within the promoter region o f the mouse connexin 43 gene. J Biol Chem. 270:3863-8. Chicoine, M . R . , and D . L . Silbergeld. 1995. Assessment o f brain tumor cell motility in vivo and in vitro. JNeurosurg. 82:615-22. Darrow, B . J . , J . G . Laing, P . D . Lampe, J.E. Saffitz, and E . C . Beyer. 1995. Expression o f multiple connexins in cultured neonatal rat ventricular myocytes. Circ Res. 76:381-7. Das Sarma, J., R . A . Meyer, F. Wang, V . Abraham, C W . L o , and M . K o v a l . 2001. Multimeric connexin interactions prior to the trans-Golgi network. J Cell Sci. 114:4013-24. Destexhe, A . , and T.J. Sejnowski. 2003. Interactions between membrane conductances underlying thalamocortical slow-wave oscillations. Physiol Rev. 83:1401-53.  97 Dilber, M . S . , M . R . Abedi, B . Christensson, B . Bjorkstrand, G . M . Kidder, C . C . Naus, G . Gahrton, and C.I. Smith. 1997. Gap junctions promote the bystander effect o f herpes simplex virus thymidine kinase in vivo. Cancer Res. 57:1523-8. Eden, S., R. Rohatgi, A . V . Podtelejnikov, M . Mann, and M . W . Kirschner. 2002. Mechanism o f regulation o f W A V E 1-induced actin nucleation by R a c l and Nek. Nature. 418:790-3. Elfgang, C , R. Eckert, H . Lichtenberg-Frate, A . Butterweck, O. Traub, R . A . K l e i n , D . F . Hulser, and K . Willecke. 1995. Specific permeability and selective formation o f gap junction channels in connexin-transfected H e L a cells. J Cell Biol. 129:805-17. Elshami, A . A . , A . Saavedra, H . Zhang, J.C. Kucharczuk, D . C . Spray, G.I. Fishman, K . M . A m i n , L . R . Kaiser, and S . M . Albelda. 1996. Gap junctions play a role in the 'bystander effect' o f the herpes simplex virus thymidine kinase/ganciclovir system i n vitro. Gene Ther. 3:8592. Enomoto, T., Y . Sasaki, Y . Shiba, Y . Kanno, and H . Yamasaki. 1981. Tumor promoters cause a rapid and reversible inhibition o f the formation and maintenance o f electrical cell coupling in culture. Proc Natl Acad Sci USA. 78:5628-32. Ewart, J.L., M . F . Cohen, R . A . Meyer, G . Y . Huang, A . Wessels, R . G . Gourdie, A . J . Chin, S . M . Park, B . O . Lazatin, S. Villabon, and C W . L o . 1997. Heart and neural tube defects in transgenic mice overexpressing the Cx43 gap junction gene. Development. 124:1281-92. Farooqui, R., and G . Fenteany. 2005. Multiple rows o f cells behind an epithelial wound edge extend cryptic lamellipodia to collectively drive cell-sheet movement. J Cell Sci. 118:5163. Fenner, W . R . , and J . A . Hass. 1989. Mechanisms o f seizure disorders. Probl Vet Med. 1:501-15. Fentiman, I.S., J. Hurst, R . L . Ceriani, and J. Taylor-Papadimitriou. 1979. Junctional intercellular communication pattern o f cultured human breast cancer cells. Cancer Res. 39:4739-43. Fernstrom, M . J . , L . D . Koffler, G . Abou-Rjaily, P . D . Boucher, D . S . Shewach, and R.J. Ruch. 2002. Neoplastic reversal o f human ovarian carcinoma cells transfected with connexin43. Exp Mol Pathol. 73:54-60. Fishman, G.I., D . C . Spray, and L . A . Leinwand. 1990. Molecular characterization and functional expression o f the human cardiac gap junction channel. J Cell Biol. I l l :589-98. Frenzel, E . M . , and R . G . Johnson. 1996. Gap junction formation between cultured embryonic lens cells is inhibited by antibody to N-cadherin. Dev Biol. 179:1-16. Fu, C.T., J.F. Bechberger, M . A . Ozog, B . Perbal, and C . C . Naus. 2004. C C N 3 ( N O V ) interacts with connexin43 in C 6 glioma cells: possible mechanism o f connexin-mediated growth suppression. J Biol Chem. 279:36943-50. Geimonen, E . , W . Jiang, M . A l i , G.I. Fishman, R . E . Garfield, and J. Andersen. 1996. Activation of protein kinase C in human uterine smooth muscle induces connexin-43 gene transcription through an A P - 1 site in the promoter sequence. J Biol Chem. 271:23667-74. Gellhaus, A . , X . Dong, S. Propson, K . Maass, L . Klein-Hitpass, M . Kibschull, O. Traub, K . Willecke, B . Perbal, S.J. L y e , and E . Winterhager. 2004. Connexin43 interacts with N O V : a possible mechanism for negative regulation o f cell growth in choriocarcinoma cells. JBiol Chem. 279:36931-42. Giepmans, B . N . 2004. Gap junctions and connexin-interacting proteins. Cardiovasc Res. 62:23345. Giepmans, B . N . , I. Verlaan, T. Hengeveld, H . Janssen, J. Calafat, M . M . Falk, and W . H . Moolenaar. 2001. Gap junction protein connexin-43 interacts directly with microtubules. CurrBiol. 11:1364-8. Gittens, J.E., and G . M . Kidder. 2005. Differential contributions o f connexin37 and connexin43 to oogenesis revealed in chimeric reaggregated mouse ovaries. J Cell Sci. 118:5071-8.  98 Gittens, J.E., A . A . M h a w i , D . Lidington, Y . Ouellette, and G . M . Kidder. 2003. Functional analysis o f gap junctions in ovarian granulosa cells: distinct role for connexin43 in early stages o f folliculogenesis. Am J Physiol Cell Physiol. 284:C880-7. Goldberg, G.S., J.F. Bechberger, and C . C . Naus. 1995. A pre-loading method o f evaluating gap junctional communication by fluorescent dye transfer. Biotechniques. 18:490-7. Goldberg, G.S., J.F. Bechberger, Y . Tajima, M . Merritt, Y . Omori, M . A . Gawinowicz, R. Narayanan, Y . Tan, Y . Sanai, H . Yamasaki, C . C . Naus, H . Tsuda, and B . J . Nicholson. 2000. Connexin43 suppresses M F G - E 8 while inducing contact growth inhibition o f glioma cells. Cancer Res. 60:6018-26. Goodenough, D . A . , J . A . Goliger, and D . L . Paul. 1996. Connexins, connexons, and intercellular communication. Annu Rev Biochem. 65:475-502. Graeber, S.H., and D . F . Hulser. 1998. Connexin transfection induces invasive properties in H e L a cells. Exp Cell Res. 243:142-9. Grobben, B . , P.P. De Deyn, and H . Siegers. 2002. Rat C 6 glioma as experimental model system for the study o f glioblastoma growth and invasion. Cell Tissue Res. 310:257-70. Grosheva, I., M . Shtutman, M . Elbaum, and A . D . Bershadsky. 2001. p l 2 0 catenin affects cell motility via modulation o f activity o f Rho-family GTPases: a link between cell-cell contact formation and regulation o f cell locomotion. J Cell Sci. 114:695-707. Gumbiner, B . M . , and P . D . McCrea. 1993. Catenins as mediators o f the cytoplasmic functions o f cadherins. J Cell Sci Suppl. 17:155-8. Gupta, N . , H . Wang, T . L . M c L e o d , C . C . Naus, S. Kyurkchiev, S. A d v a n i , J. Y u , B . Perbal, and R . R . Weichselbaum. 2001. Inhibition o f glioma cell growth and tumorigenic potential by C C N 3 ( N O V ) . Mol Pathol. 54:293-9. Gutmann, D . H . , and M . Giovannini. 2002. Mouse models o f neurofibromatosis 1 and 2. Neoplasia. 4:279-90. Hatta, K . , T.S. Okada, and M . Takeichi. 1985. A monoclonal antibody disrupting calciumdependent cell-cell adhesion o f brain tissues: possible role o f its target antigen in animal pattern formation. Proc Natl Acad Sci USA. 82:2789-93. Haubrich, S., H . J . Schwarz, F. Bukauskas, H . Lichtenberg-Frate, O. Traub, R. Weingart, and K . Willecke. 1996. Incompatibility o f connexin 40 and 43 Hemichannels i n gap junctions between mammalian cells is determined by intracellular domains. Mol Biol Cell. 7:19952006. Hayashi, T., K . Nomata, C . C . Chang, R . J . Ruch, and J.E. Trosko. 1998. Cooperative effects o f vmyc and c-Ha-ras oncogenes on gap junctional intercellular communication and tumorigenicity in rat liver epithelial cells. Cancer Lett. 128:145-54. Herve, J . C , N . Bourmeyster, and D . Sarrouilhe. 2004a. Diversity in protein-protein interactions of connexins: emerging roles. Biochim Biophys Acta. 1662:22-41. Herve, J . C , I. Plaisance, J. Loncarek, F. Duthe, and D . Sarrouilhe. 2004b. Is the junctional uncoupling elicited in rat ventricular myocytes by some dephosphorylation treatments due to changes in the phosphorylation status o f Cx43? Eur Biophys J. 33:201-10. Herve, J . C , and D . Sarrouilhe. 2006. Protein phosphatase modulation o f the intercellular junctional communication: importance in cardiac myocytes. Prog Biophys Mol Biol. 90:225-48. Hoffman, A . G . , R . C . Burghardt, R. Tilley, and N . Auersperg. 1993. A n in vitro model o f ovarian epithelial carcinogenesis: changes in cell-cell communication and adhesion occurring during neoplastic progression. Int J Cancer. 54:828-38. Homma, N . , J.L. Alvarado, W . Coombs, K . Stergiopoulos, S . M . Taffet, A . F . L a u , and M . Delmar. 1998. A particle-receptor model for the insulin-induced closure o f connexin43 channels. CircRes. 83:27-32.  99 Hou, J . C , S. Shigematsu, H . C . Crawford, P . Z . Anastasiadis, and J.E. Pessin. 2006. Dual Regulation o f Rho and Rac by p i 2 0 Catenin Controls Adipocyte Plasma Membrane Trafficking. J Biol Chem. 281:23307-12. Huang, G . Y . , E . S . Cooper, K . Waldo, M . L . Kirby, N . B . Gilula, and C . W . L o . 1998a. Gap junction-mediated cell-cell communication modulates mouse neural crest migration. J Cell Biol. 143:1725-34. Huang, G . Y . , A . Wessels, B . R . Smith, K . K . Linask, J . L . Ewart, and C . W . L o . 1998b. Alteration in connexin 43 gap junction gene dosage impairs conotruncal heart development. Dev Biol. 198:32-44. Huang, R . P . , Y . Fan, M . Z . Hossain, A . Peng, Z . L . Zeng, and A . L . Boynton. 1998c. Reversion o f the neoplastic phenotype o f human glioblastoma cells by connexin 43 (cx43). Cancer Res. 58:5089-96. Huang, R . P . , M . Z . Hossain, A . Sehgal, and A . L . Boynton. 1999. Reduced connexin43 expression in high-grade human brain glioma cells. J Surg Oncol. 70:21-4. Huard, C , N . Druesne, D . Guyonnet, M . Thomas, A . Pagniez, A . M . L e B o n , P. Martel, and C . Chaumontet. 2004. Diallyl disulfide ( D A D S ) enhances gap-junctional intercellular communication by both direct and indirect mechanisms in rat liver cells. Carcinogenesis. 25:91-8. Huttenlocher, A . , R . R . Sandborg, and A . F . Horwitz. 1995. Adhesion in cell migration. Curr Opin Cell Biol. 1:691'-706. Iacobas, D . A . , E . Scemes, and D . C . Spray. 2004. Gene expression alterations in connexin null mice extend beyond the gap junction. Neurochem Int. 45:243-50. Janmey, P . A . , and T.P. Stossel. 1987. Modulation o f gelsolin function by phosphatidylinositol 4,5-bisphosphate. Nature. 325:362-4. Jongen, W . M . , D . J . Fitzgerald, M . Asamoto, C . Piccoli, T.J. Slaga, D . Gros, M . Takeichi, and H . Yamasaki. 1991. Regulation o f connexin 43-mediated gap junctional intercellular communication by Ca2+ in mouse epidermal cells is controlled by E-cadherin. J Cell Biol. 114:545-55. Jou, Y . S . , B . Layhe, D . F . Matesic, C . C . Chang, A . W . de Feijter, L . L o c k w o o d , C . W . Welsch, J.E. Klaunig, and J.E. Trosko. 1995. Inhibition o f gap junctional intercellular communication and malignant transformation o f rat liver epithelial cells by neu oncogene. Carcinogenesis. 16:311-7. Jou, Y . S . , D . Matesic, E . Dupont, S . C L u , H . L . Rupp, B . V . Madhukar, S . Y . O h , J.E. Trosko, and C . C . Chang. 1993. Restoration o f gap-junctional intercellular communication in a communication-deficient rat liver cell mutant by transfection with connexin 43 c D N A . Mol Carcinog. 8:234-44. K a l i m i , G . H . , L . L . Hampton, J.E. Trosko, S.S. Thorgeirsson, and A . C . Huggett. 1992. Homologous and heterologous gap-junctional intercellular communication in v-raf-, vmyc-, and v-raf/v-myc-transduced rat liver epithelial cell lines. Mol Carcinog. 5:301-10. Kamibayashi, Y . , Y . Oyamada, M . M o r i , and M . Oyamada. 1995. Aberrant expression o f gap junction proteins (connexins) is associated with tumor progression during multistage mouse skin carcinogenesis in vivo. Carcinogenesis. 16:1287-97. Kapoor, P., M . M . Saunders, Z . L i , Z . Zhou, N . Sheaffer, E . L . Kunze, R . S . Samant, D . R . Welch, and H . J . Donahue. 2004. Breast cancer metastatic potential: correlation with increased heterotypic gap junctional intercellular communication between breast cancer cells and osteoblastic cells. Int J Cancer. 111:693-7. Kelsell, D . P . , J. Dunlop, and M . B . Hodgins. 2001. Human diseases: clues to cracking the connexin code? Trends Cell Biol. 11:2-6.  100 Kelsell, D . P . , J. Dunlop, H . P . Stevens, N . J . Lench, J . N . Liang, G . Parry, R . F . Mueller, and L M . Leigh. 1997. Connexin 26 mutations in hereditary non-syndromic sensorineural deafness. Nature. 387:80-3. K i n g , T.J., L . H . Fukushima, Y . Yasui, P . D . Lampe, and J.S. Bertram. 2002. Inducible expression of the gap junction protein connexin43 decreases the neoplastic potential o f HT-1080 human fibrosarcoma cells in vitro and in vivo. Mol Carcinog. 35:29-41. Klingelhofer, J., R . B . Troyanovsky, O . Y . Laur, and S. Troyanovsky. 2000. Amino-terminal domain o f classic cadherins determines the specificity o f the adhesive interactions. J Cell Sci. 113 ( P t 16):2829-36. Knott, J.C., R. Mahesparan, I. Garcia-Cabrera, B . Bolge Tysnes, K . Edvardsen, G . O . Ness, S. M o r k , M . Lund-Johansen, and R. Bjerkvig. 1998. Stimulation o f extracellular matrix components in the normal brain by invading glioma cells. Int J Cancer. 75:864-72. Kobielak, A . , H . A . Pasolli, and E . Fuchs. 2004. Mammalian formin-1 participates in adherens junctions and polymerization o f linear actin cables. Nat Cell Biol. 6:21-30. Korones, D . N . , J. Padowski, B . A . Factor, and L . S . Constine. 2003. D o children with optic pathway tumors have an increased frequency o f other central nervous system tumors? Neuro-oncol. 5:116-20. Krause, M . , E . W . Dent, J.E. Bear, J.J. Loureiro, and F . B . Gertler. 2003. E n a / V A S P proteins: regulators o f the actin cytoskeleton and cell migration. Annu Rev Cell Dev Biol. 19:54164. Kren, B . T . , N . M . Kumar, S.Q. Wang, N . B . Gilula, and C.J. Steer. 1993. Differential regulation o f multiple gap junction transcripts and proteins during rat liver regeneration. J Cell Biol. 123:707-18. Kretz, M . , C . Euwens, S. Hombach, D . Eckardt, B . Teubner, O. Traub, K . Willecke, and T. Ott. 2003. Altered connexin expression and wound healing in the epidermis o f connexindeficient mice. J Cell Sci. 116:3443-52. Laird, D . W . , K . L . Puranam, and J.P. Revel. 1991. Turnover and phosphorylation dynamics o f connexin43 gap junction protein in cultured cardiac myocytes. Biochem J. 273(Pt 1):6772. Lampe, P.D., and A . F . L a u . 2000. Regulation o f gap junctions by phosphorylation o f connexins. Arch Biochem Biophys. 384:205-15. Lampe, P.D., and A . F . L a u . 2004. The effects o f connexin phosphorylation on gap junctional communication. Int J Biochem Cell Biol. 36:1171-86. Lampe, P.D., Q. Q i u , R . A . Meyer, E . M . TenBroek, T . F . Walseth, T . A . Starich, H . L . Grunenwald, and R . G . Johnson. 2001. Gap junction assembly: PTX-sensitive G proteins regulate the distribution o f connexin43 within cells. Am J Physiol Cell Physiol. 28LC1211-22. Lauf, U . , B . N . Giepmans, P. Lopez, S. Braconnot, S.C. Chen, and M . M . Falk. 2002. Dynamic trafficking and delivery o f connexons to the plasma membrane and accretion to gap junctions in living cells. Proc Natl Acad Sci USA. 99:10446-51. Lauffenburger, D . A . , and A . F . Horwitz. 1996. C e l l migration: a physically integrated molecular process. Cell. 84:359-69. Laurent, M . , C . Martinerie, H . Thibout, M . P . Hoffman, F. Verrecchia, Y . L e Bouc, A . Mauviel, and H . K . Kleinman. 2003. N O V H increases M M P 3 expression and cell migration in glioblastoma cells via a PDGFR-alpha-dependent mechanism. Faseb J. 17:1919-21. Le, D . M . , A . Besson, D . K . Fogg, K . S . Choi, D . M . Waisman, C . G . Goodyer, B . Rewcastle, and V . W . Y o n g . 2003. Exploitation o f astrocytes by glioma cells to facilitate invasiveness: a mechanism involving matrix metalloproteinase-2 and the urokinase-type plasminogen activator-plasmin cascade. J Neurosci. 23:4034-43.  101 Li, F., and H.N. Higgs. 2003. The mouse Formin mDial is a potent actin nucleation factor regulated by autoinhibition. Curr Biol. 13:1335-40. Li, W.E., K. Waldo, K.L. Linask, T. Chen, A. Wessels, M.S. Parmacek, M.L. Kirby, and C.W. Lo. 2002. An essential role for connexin43 gap junctions in mouse coronary artery development. Development. 129:2031-42. Lin, C.G., S.J. Leu, N . Chen, C M . Tebeau, S.X. Lin, C.Y. Yeung, and L.F. Lau. 2003a. CCN3 (NOV) is a novel angiogenic regulator of the CCN protein family. J Biol Chem. 278:24200-8. Lin, J.H., T. Takano, M.L. Cotrina, G. Arcuino, J. Kang, S. Liu, Q. Gao, L. Jiang, F. Li, H. Lichtenberg-Frate, S. Haubrich, K. Willecke, S.A. Goldman, and M . Nedergaard. 2002. Connexin 43 enhances the adhesivity and mediates the invasion of malignant glioma cells. J Neurosci. 22:4302-11. Lin, J.H., J. Yang, S. Liu, T. Takano, X. Wang, Q. Gao, K. Willecke, and M . Nedergaard. 2003b. Connexin mediates gap junction-independent resistance to cellular injury. J Neurosci. 23:430-41. Lo, C.W., and N.B. Gilula. 1979. Gap junctional communication in the preimplantation mouse embryo. Cell. 18:399-409. Loewenstein, W.R., and Y. Kanno. 1966. Intercellular communication and the control of tissue growth: lack of communication between cancer cells. Nature. 209:1248-9. Louis, D.N., E.C Holland, and J.G. Cairncross. 2001. Glioma classification: a molecular reappraisal. Am J Pathol. 159:779-86. Maheshwari, G., and D.A. Lauffenburger. 1998. Deconstructing (and reconstructing) cell migration. Microsc Res Tech. 43:358-68. Majesky, M.W. 1994. Neointima formation after acute vascular injury. Role of counteradhesive extracellular matrix proteins. Tex Heart Inst J. 21:78-85. Manthey, D., F. Bukauskas, C G . Lee, C A . Kozak, and K. Willecke. 1999. Molecular cloning and functional expression of the mouse gap junction gene connexin-57 in human HeLa cells. JBiol Chem. 274:14716-23. Mao, A.J., J. Bechberger, D. Lidington, J. Galipeau, D.W. Laird, and C.C. Naus. 2000. Neuronal differentiation and growth control of neuro-2a cells after retroviral gene delivery of connexin43. JBiol Chem. 275:34407-14. McLeod, T.L., J.F. Bechberger, and C.C. Naus. 2001. Determination of a potential role of the CCN family of growth regulators in connexin43 transfected C6 glioma cells. Cell Commun Adhes. 8:441-5.  Mercola, M., and M . Levin. 2001. Left-right asymmetry determination in vertebrates. Annu Rev Cell Dev Biol. 17:779-805. Mesnil, M., S. Crespin, J.L. Avanzo, and M.L. Zaidan-Dagli. 2005. Defective gap junctional intercellular communication in the carcinogenic process. Biochim Biophys Acta. 1719:125-45. Moorby, C D . 2000. A connexin 43 mutant lacking the carboxyl cytoplasmic domain inhibits both growth and motility of mouse 3T3 fibroblasts. Mol Carcinog. 28:23-30. Moorby, C D . , and E. Gherardi. 1999. Expression of a Cx43 deletion mutant in 3T3 A31 fibroblasts prevents PDGF-induced inhibition of cell communication and suppresses cell growth. Exp Cell Res. 249:367-76. Moreno, A.P., M . Chanson, S. Elenes, J. Anumonwo, I. Scerri, H. Gu, S.M. Taffet, and M . Delmar. 2002. Role of the carboxyl terminal of connexin43 in transjunctional fast voltage gating. Circ Res. 90:450-7. Moss, A.R. 1985. Occupational exposure and brain tumors. J Toxicol Environ Health. 16:70311.  102 Mullins, R . D . , J . A . Heuser, and T . D . Pollard. 1998. The interaction o f Arp2/3 complex with actin: nucleation, high affinity pointed end capping, and formation o f branching networks of filaments. Proc Natl Acad Sci USA. 95:6181-6. M u s i l , L . S . , and D . A . Goodenough. 1993. Multisubunit assembly o f an integral plasma membrane channel protein, gap junction connexin43, occurs after exit from the E R . Cell. 74:1065-77. M u s i l , L . S . , A . C . L e , J . K . VanSlyke, and L . M . Roberts. 2000. Regulation o f connexin degradation as a mechanism to increase gap junction assembly and function. JBiol Chem. 275:25207-15. Nagafuchi, A . 2001. Molecular architecture o f adherens junctions. Curr Opin Cell Biol. 13:6003. Naus, C . C . , S.L. Bond, J.F. Bechberger, and W . Rushlow. 2000. Identification o f genes differentially expressed in C 6 glioma cells transfected with connexin43. Brain Res Brain Res Rev. 32:259-66. Naus, C . C . , K . Elisevich, D . Zhu, D . J . Belliveau, and R . F . D e l Maestro. 1992. In vivo growth o f C6 glioma cells transfected with connexin43 c D N A . Cancer Res. 52:4208-13. Naves, M . M . , E . R . Silveira, M . L . Dagli, and F . S . Moreno. 2001. Effects o f beta-carotene and vitamin A on oval cell proliferation and connexin 43 expression during hepatic differentiation in the rat(l). JNutr Biochem. 12:685-692. Ohgaki, H . , and P. Kleihues. 2005a. Epidemiology and etiology o f gliomas. Acta Neuropathol (Berl). 109:93-108. Ohgaki, H . , and P. Kleihues. 2005b. Population-based studies on incidence, survival rates, and genetic alterations in astrocytic and oligodendroglial gliomas. J Neuropathol Exp Neurol. 64:479-89. Oliveira, R., C . Christov, J.S. Guillamo, S. de Bouard, S. Palfi, L . Venance, M . Tardy, and M . Peschanski. 2005. Contribution o f gap junctional communication between tumor cells and astroglia to the invasion o f the brain parenchyma by human glioblastomas. BMC Cell Biol. 6:7. Omori, Y . , and H . Yamasaki. 1998. Mutated connexin43 proteins inhibit rat glioma cell growth suppression mediated by wild-type connexin43 in a dominant-negative manner. Int J Cancer. 78:446-53. Ozog, M . A . , J.F. Bechberger, and C . C . Naus. 2002. Ciliary neurotrophic factor ( C N T F ) in combination with its soluble receptor (CNTFRalpha) increases connexin43 expression and suppresses growth o f C 6 glioma cells. Cancer Res. 62:3544-8. Ozog, M . A . , S . M . Bernier, D . C . Bates, B . Chatterjee, C W . L o , and C . C . Naus. 2004. The complex o f ciliary neurotrophic factor-ciliary neurotrophic factor receptor alpha upregulates connexin43 and intercellular coupling in astrocytes via the Janus tyrosine kinase/signal transducer and activator o f transcription pathway. Mol Biol Cell. 15:476174. Paznekas, W . A . , S.A. Boyadjiev, R . E . Shapiro, O. Daniels, B . W o l l n i k , C E . Keegan, J.W. Innis, M . B . Dinulos, C . Christian, M . C . Hannibal, and E . W . Jabs. 2003. Connexin 43 (GJA1) mutations cause the pleiotropic phenotype o f oculodentodigital dysplasia. Am J Hum Genet. 72:408-18. Pecina-Slaus, N . 2003. Tumor suppressor gene E-cadherin and its role in normal and malignant cells. Cancer Cell Int. 3:17. Peitsch, W . K . , I. Hofmann, S. Pratzel, C . Grund, C . Kuhn, I. M o l l , L . Langbein, and W . W . Franke. 2001. Drebrin particles: components in the ensemble o f proteins regulating actin dynamics o f lamellipodia and filopodia. Eur J Cell Biol. 80:567-79.  103 Perez Velazquez, J.L., and P . L . Carlen. 2000. Gap junctions, synchrony and seizures. Trends Neurosci. 23:68-74. Pollmann, M . A . , Q. Shao, D . W . Laird, and M . Sandig. 2005. Connexin 43 mediated gap junctional communication enhances breast tumor cell diapedesis in culture. Breast Cancer Res. 7:R522-34. Pu, P., Z . X i a , S. Y u , and Q. Huang. 2004. Altered expression o f Cx43 in astrocytic tumors. Clin Neurol Neurosurg. 107:49-54. Qiu, C , P. Coutinho, S. Frank, S. Franke, L . Y . L a w , P. Martin, C R . Green, and D . L . Becker. 2003. Targeting connexin43 expression accelerates the rate o f wound repair. Curr Biol. 13:1697-703. Reaume, A . G . , P . A . de Sousa, S. Kulkarni, B . L . Langille, D . Z h u , T . C . Davies, S . C Juneja, G . M . Kidder, and J. Rossant. 1995. Cardiac malformation in neonatal mice lacking connexin43. Science. 267:1831-4. Richard, G . , F. Rouan, C . E . Willoughby, N . Brown, P. Chung, M . Ryynanen, E . W . Jabs, S.J. Bale, J.J. DiGiovanna, J. Uitto, and L . Russell. 2002. Missense mutations in G J B 2 encoding connexin-26 cause the ectodermal dysplasia keratitis-ichthyosis-deafness syndrome. Am J Hum Genet. 70:1341-8. Rosolen, A . , E . Frascella, C . di Francesco, A . Todesco, M . Petrone, M . Mehtali, F. Zacchello, L . Zanesco, and M . Scarpa. 1998. In vitro and in vivo antitumor effects o f retro virusmediated herpes simplex thymidine kinase gene-transfer in human medulloblastoma. Gene Ther. 5:113-20. Saez, J . C , V . M . Berthoud, M . C . Branes, A . D . Martinez, and E . C Beyer. 2003. Plasma membrane channels formed by connexins: their regulation and functions. Physiol Rev. 83:1359-400. Sato, Y . , and D . B . Rifkin. 1988. Autocrine activities o f basic fibroblast growth factor: regulation of endothelial cell movement, plasminogen activator synthesis, and D N A synthesis: J Cell Biol. 107:1199-205. Schiavi, A . , A . Hudder, and R. Werner. 1999. Connexin43 m R N A contains a functional internal ribosome entry site. FEBS Lett. 464:118-22. Seki, A . , W . Coombs, S . M . Taffet, and M . Delmar. 2004. Loss o f electrical communication, but not plaque formation, after mutations in the cytoplasmic loop o f connexin43. Heart Rhythm. 1:227-33. Shao, Q., H . Wang, E . McLachlan, G.I. Veitch, and D . W . Laird. 2005. Down-regulation o f Cx43 by retroviral delivery o f small interfering R N A promotes an aggressive breast cancer cell phenotype. Cancer Res. 65:2705-11. Shiels, A . , D . Mackay, A . Ionides, V . Berry, A . Moore, and S. Bhattacharya. 1998. A missense mutation in the human connexin50 gene ( G J A 8 ) underlies autosomal dominant "zonular pulverulent" cataract, on chromosome l q . Am J Hum Genet. 62:526-32. Shima, K . , T. Muramatsu, Y . A b i k o , Y . Yamaoka, H . Sasaki, and M . Shimono. 2006. Connexin 43 transfection in basaloid squamous cell carcinoma cells. Oncol Rep. 16:285-8. Shinoura, N . , L . Chen, M . A . W a n i , Y . G . K i m , J.J. Larson, R . E . Warnick, M . Simon, A . G . Menon, W . L . B i , and P.J. Stambrook. 1996. Protein and messenger R N A expression o f connexin43 in astrocytomas: implications in brain tumor gene therapy. J Neurosurg. 84:839-45; discussion 846. Simon, A . M . , and D . A . Goodenough. 1998. Diverse functions o f vertebrate gap junctions. Trends Cell Biol. 8:477-83. Simon, A . M . , D . A . Goodenough, E . L i , and D . L . Paul. 1997. Female infertility in mice lacking connexin 37. Nature. 385:525-9.  104 Singh, S.K., I.D. Clarke, M . Terasaki, V . E . Bonn, C . Hawkins, J. Squire, and P . B . Dirks. 2003. Identification o f a cancer stem cell in human brain tumors. Cancer Res. 63:5821-8. Soroceanu, L . , T.J. Manning, Jr., and H . Sontheimer. 2001. Reduced expression o f connexin-43 and functional gap junction coupling in human gliomas. Glia. 33:107-17. Spray, D . C , and M . V . Bennett. 1985. Physiology and pharmacology o f gap junctions. Annu Rev Physiol. 47:281-303. Sullivan, R., C . Ruangvoravat, D . Joo, J. Morgan, B . L . Wang, X . K . Wang, and C W . L o . 1993. Structure, sequence and expression o f the mouse Cx43 gene encoding connexin 43. Gene. 130:191-9. Swanson, K . R . , C . Bridge, J.D. Murray, and E . C . A l v o r d , Jr. 2003. Virtual and real brain tumors: using mathematical modeling to quantify glioma growth and invasion. J Neurol Sci. 216:1-10. Swierenga, S.H., and H . Yamasaki. 1992. Performance o f tests for cell transformation and gapjunction intercellular communication for detecting nongenotoxic carcinogenic activity. IARC Sci Publ: 165-93. Takeichi, M . 1988. The cadherins: cell-cell adhesion molecules controlling animal morphogenesis. Development. 102:639-55. Trosko, J.E., and R.J. Ruch. 2002. Gap junctions as targets for cancer chemoprevention and chemotherapy. Curr Drug Targets. 3:465-82. Valiunas, V . , Y . Y . Polosina, H . Miller, L A . Potapova, L . Valiuniene, S. Doronin, R . T . Mathias, R . B . Robinson, M . R . Rosen, L S . Cohen, and P.R. Brink. 2005. Connexin-specific cell-tocell transfer o f short interfering R N A by gap junctions. J Physiol. 568:459-68. van der Heyden, M . A . , M . B . Rook, M . M . Hermans, G . Rijksen, J. Boonstra, L . H . Defize, and O . H . Destree. 1998. Identification o f connexin43 as a functional target for Wnt signalling. J Cell Sci. 111 ( P t 12): 1741-9. van R o y , F . M . , and P . D . McCrea. 2005. A role for Kaiso-pl20ctn complexes in cancer? Nat Rev Cancer. 5:956-64. VanSlyke, J.a.M., L . 2005. Exogenous overexpression o f connexins induces premature assembly of connexons in the E R . In I G J conference. V o l . Session 1: Life Cycle o f Connexins. Veitch, G.I., J.E. Gittens, Q. Shao, D . W . Laird, and G . M . Kidder. 2004. Selective assembly o f connexin37 into heterocellular gap junctions at the oocyte/granulosa cell interface. J Cell Sci. 117:2699-707. Vicente-Manzanares, M . , D . J . Webb, and A . R . Horwitz. 2005. C e l l migration at a glance. J Cell Sci. 118:4917-9. Vigetti, D . , P. Moretto, M . V i o l a , A . Genasetti, M . R i z z i , E . Karousou, F . Pallotti, G . De Luca, and A . Passi. 2006. Matrix metalloproteinase 2 and tissue inhibitors o f metalloproteinases regulate human aortic smooth muscle cell migration during i n vitro aging. Faseb J. 20:1118-30. Vukicevic, S., H . K . Kleinman, F.P. Luyten, A . B . Roberts, N . S . Roche, and A . H . Reddi. 1992. Identification o f multiple active growth factors in basement membrane Matrigel suggests caution in interpretation o f cellular activity related to extracellular matrix components. Exp Cell Res. 202:1-8. Wang, Y . , and B . Rose. 1997. A n inhibition o f gap-junctional communication by cadherins. J Cell Sci. 110 ( P t 3):301-9. W e i , C.J., R. Francis, X . X u , and C W . L o . 2005. Connexin43 associated with an N-cadherincontaining multiprotein complex is required for gap junction formation in N I H 3 T 3 cells. J Biol Chem. 280:19925-36. Werbowetski, T., R. Bjerkvig, and R . F . D e l Maestro. 2004. Evidence for a secreted chemorepellent that directs glioma cell invasion. JNeurobiol. 60:71-88.  105 Wheelock, M . J . , and K . R . Johnson. 2003. Cadherin-mediated cellular signaling. Curr Opin Cell Biol. 15:509-14. White, T . W . , D . L . Paul, D . A . Goodenough, and R. Bruzzone. 1995. Functional analysis o f selective interactions among rodent connexins. Mol Biol Cell. 6:459-70. Wilkinson, P . C . 1998. Assays o f leukocyte locomotion and chemotaxis. J Immunol Methods. 216:139-53. Willecke, K . , J. Eiberger, J. Degen, D . Eckardt, A . Romualdi, M . Guldenagel, U . Deutsch, and G . Sohl. 2002. Structural and functional diversity o f connexin genes in the mouse and human genome. Biol Chem. 383:725-37. Woods, S.A., E . Marmor, M . Feldkamp, N . Lau, A . J . A p i c e l l i , G . Boss, D . H . Gutmann, and A . Guha. 2002. Aberrant G protein signaling in nervous system tumors. J Neurosurg. 97:627-42. W u , J.C., R . Y . Tsai, and T . H . Chung. 2003. Role o f catenins in the development o f gap junctions in rat cardiomyocytes. J Cell Biochem. 88:823-35. X u , G . , A . W . Craig, P. Greer, M . Miller, P . Z . Anastasiadis, J. L i l i e n , and J. Balsamo. 2004. Continuous association o f cadherin with beta-catenin requires the non-receptor tyrosinekinase Fer.JCell Sci. 117:3207-19. X u , X . , R. Francis, C.J. W e i , K . L . Linask, and C . W . L o . 2006. Connexin 43-mediated modulation o f polarized cell movement and the directional migration o f cardiac neural crest cells. Development. X u , X . , W . E . L i , G . Y . Huang, R. Meyer, T. Chen, Y . L u o , M . P . Thomas, G . L . Radice, and C . W . L o . 2001. Modulation o f mouse neural crest cell motility b y N-cadherin and connexin 43 gap junctions. J Cell Biol. 154:217-30. Y u , W . , G . Dahl, and R. Werner. 1994. The connexin43 gene is responsive to oestrogen. Proc Biol Sci. 255:125-32. Zagzag, D . , B . Shiff, G.I. Jallo, M . A . Greco, C . Blanco, H . Cohen, J. Hukin, J . C . A l l e n , and D . R . Friedlander. 2002. Tenascin-C promotes microvascular cell migration and phosphorylation o f focal adhesion kinase. Cancer Res. 62:2660-8. Zhang, W . , C . N w a g w u , D . M . L e , V . W . Y o n g , H . Song, and W . T . Couldwell. 2003a. Increased invasive capacity o f connexin43-overexpressing malignant glioma cells. J Neurosurg. 99:1039-46. Zhang, Y . W . , M . Kaneda, and I. Morita. 2003b. The gap junction-independent tumorsuppressing effect o f connexin 43. JBiol Chem. 278:44852-6. Zhou, L . , E . M . Kasperek, and B . J . Nicholson. 1999. Dissection o f the molecular basis o f pp60(vsrc) induced gating o f connexin 43 gap junction channels. J Cell Biol. 144:1033-45. Zhu, D . , S. Caveney, G . M . Kidder, and C . C . Naus. 1991. Transfection o f C 6 glioma cells with connexin 43 c D N A : analysis o f expression, intercellular coupling, and cell proliferation. Proc Natl Acad Sci USA. 88:1883-7. Zhu, D . , G . M . Kidder, S. Caveney, and C . C . Naus. 1992. Growth retardation in glioma cells cocultured with cells overexpressing a gap junction protein. Proc Natl Acad Sci USA. 89:10218-21.  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items