INFORMATION TO USERS

This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adverselyaffect reproduction.

In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note willindicate the deletion.

Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand corner and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the hook.

Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.

V·M·I University Microfilms International A Bell & Howell Information Company 300 North Zeeb Road. Ann Arbor. M148106-1346 USA 313:761-4700 800/521-0600

Order Number 9325035

Characterization of epidermal growth factor-induced disruption of gap junctional intercellular communication and connexin43

Kanemitsu, Martha Yasuko, Ph.D. University of Hawaii, 1993

D·M·I 300N. ZeebRd. AnnArbor,MI48106

CHARACTERIZATION OF EPIDERMAL GROWTH FACTOR-INDUCED DISRUPTION OF GAP JUNCTIONAL INTERCEllULAR COMMUNICATION AND CONNEXIN43 PHOSPHORYLATION

A DISSERTATION SUBMiTTED TO THE GRADUATE DIVISION OF THE UNIVERSITY OF HAWAII IN PARTIAL FULFilLMENT OF THE REQUIREMENTS FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY IN BIOMEDICAL SCIENCES (PHYSIOLOGY­ , MOLECULAR AND NEURO SCIENCES)

MAY 1993

BY MARTHA YASUKO KANEMITSU

DISSERTATION COMMITIEE:

MARTIN D. RAYNER, CHAIRPERSON ALAN F. LAU TOM D. HUMPHREYS PETER C. RUBEN HOWARD F. MOWER Acknowledgments

The author would like to acknowledgethe following people: Dr. Alan F. Lau for guidance and support; Linda Kuriyama for secretarial assistance; Family, friends and colleagues for their endless encouragement and support.

The author would like to acknowledge the following programs for support: Research Centers in Minorities Institutes (RCMI), Research Corporation of the University of Hawaii - Presidents Scholarship and the Cancer Research Center of Hawaii.

Sections of this dissertation have been previously published. Permission to reprint portions of the article in this dissertation has been received from Molecular of the Cell © 1992, 1993, The American Society for Cell Biology and the Graduate Division of the University of Hawaii-Manoa.

iii Abstract

Binding of a growth factor such as epidermal growth factor (EGF), to its specific cell surface receptor results in numerous intracellular responses, including increased DNA synthesis, regulation of epithelial proliferation and differentiation. The events linking EGF-induced receptor activation to these responses are still largely unknown. My work shows that EGF treatment of T51 B rat liver epithelial cells resulted in a rapid and transient disruption of gap junctional communication (gjc). EGF also stimulated the activation of its in a time-dependent manner which occurred prior to the disruption of gjc. These results suggested that EGF receptor activation may be the first of a cascade of events mediating the disruption of gjc. EGF's ability to disrupt gjc was not due to any gross alterations in gap junctional plaques or cell-cell contacts between adjacent cells. Coincident with the disruption of gjc was the serine phosphorylation of connexin43 (Cx43). In addition to this kinetic similarity, EGF's effects were dose-dependent which also suggested that a functional relationship may exist. This hypothesis was further substantiated when okadaic acid, a serine/threonine phosphatase inhibitor prevented the dephosphorylation and concomitant restoration of gjc. Therefore, it appeared that EGF­ induced disruption of gjc functionally depends on the post-translational serine phosphorylation of Cx43. While the signalling cascade(s) mediating EGF's effects has not yet been defined, down-modulation of TPA-sensitive PKC activity did not affect EGF's ability to reduce junctional permeability or phosphorylate Cx43. This suggested that TPA-sensitive PKC may not be required. EGF, in the presence or absence of chronic TPA treatment, stimulated marked increases in Cx43 phosphorylation on numerous sites iv as determined by two-dimensional tryptic phosphopeptide mapping. It is possible that Cx43 may be phosphorylated and/or regulated by multiple serine kinases. Computer-assisted sequence analysis of Cx43 identified several protein kinase phosphorylation consensus sites including two sites for mitogen-activated protein (MAP) kinase. Thus, MAP kinase may mediate, in part, EGF's ability to disrupt gjc and phosphorylate Cx43. EGF stimulated the tyrosine phosphorylation and activation of MAP kinase in a time- and dose-dependent manner where the kinetics of kinase activity corroborated its possible involvement in mediating EGF's effects. Moreover, purified MAP kinase directly phosphorylated Cx43 in vitro. Two­ dimensional tryptic phosphopeptide mapping demonstrated that the in vitro tryptic phosphopeptides represented a specific subset of the in vivo tryptic phosphopeptides produced in response to EGF following chronic TPA treatment. Therefore, EGF-induced disruption of gjc may be a functional consequence of Cx43 phosphorylation.

v Table of Contents

Acknowledgments .iii

Abstract. iv

List of Figures vii

Chapter 1: Introduction 1

Chapter 2: Materials and Methods 14

Chapter 3: EGF-induced disruption of gap junctional communication is correlated with connexin43 phosphorylation 24

Chapter 4: EGF is able to modulate gap junctional communication and connexin43 phosphorylation by a signalling pathway independent of protein kinase C 38

Chapter 5: EGF-induced disruption of gap junctional communication and connexin43 phosphorylation may be mediated by MAP kinase 50

Chapter 6: Discussion and Conclusion 62

References 71

vi List of Figures Figure Page 1.1 Schematic diagram of plaque and protein 12

3.1 EGF-induced disruption of gjc in T518 cells 29

3.2 EGF-induced receptor tyrosine kinase activation 30

3.3 Immunofluorescence antibody staining of Cx43 in T518 cells...31

3.4 EGF-induced serine phosphorylation of Cx43 33

3.5 Dose-dependence of EGF-induced disruption of gjc and Cx43 phosphorylation 35

3.6 Effect of Okadaic acid on EGF-induced disruption of gjc and Cx43 phosphorylation 36

4.1 Effect of chronic TPA treatment on EGF- and TPA-induced disruption of gjc and Cx43 phosphorylation .43

4.2 Effect of chronic TPA treatment on cytosolic and membrane- bound PKC activity in T518 cells 45

4.3 EGF-induced Cx43 phosphorylation following chronic TPA treatment occurs on serine residues .47 vii list of Figures (cont.) Figure

4.4 Two-dimensional tryptic phosphopeptide mapping of 32P-labeled Cx43 in unstimulated and EGF-treated cells: Effects of chronic TPA treatment.. .48

5.1 Kinetics and dose-dependence of EGF-stimulated mobility shift and tyrosine phosphorylation of p42 and p44 isoforms of MAP kinase 55

5.2 Kinetics and dose-dependence of EGF-induced MAP kinase activation 57

5.3 MAP kinase phosphorylates Cx43 in vitro 59

5.4 Tryptic phosphopeptides of Cx43 phosphorylated in vitro by MAP kinase comigrate with tryptic phosphopeptides of Cx43 phosphorylated in vivo in response to EGF 60

viii Chapter 1 Introduction

Structure of Gap Junctions Gap junctions were initially characterized in electron micrographs as membranous areas between cells separated by a 2 nm gap (Robertson, 1963). They have since been characterized as intercellular channels composed of hemichannels (connexons) in the plasma membrane of one cell "docking" to connexons from adjacent cells creating continuous aqueous channels. These channels allow cells to communicate both electrically and chemically. Each connexon is comprised of six protein subunits termed (Figure 1A) and each connexin protein is believed to span the membrane four times (Figure 1B). The third membrane spanning domain is believed to contribute a portion of the central core channel. The amino- and carboxy-terminal ends as well as the intracellular loop located medial to the second and third membrane spanning domains extend into the (Figure 1B). It is believed that these cytoplasmic domains of the protein are involved in regulating the permeability of gap junction channels. In addition to the membrane spanning and cytoplasmic regions, there are two extracellular loops (Figure 1B) each containing three conserved cysteine residues. These loops are thought to be involved in the "docking" process. Connexins are now recognized as a family of gap junction proteins showing evolutionary conservation and diverse expression in various tissues (Beyer et aI., 1990; Dermietzel et aI., 1990). Recent technological advances in molecular biology have facilitated the discovery of novel members of the connexin family throughout the evolutionary scale from

1 invertebrates to vertebrates. In addition to evolutionary diversity, certain connexin species may show extensive tissue distribution. For example, connexin43 is found in cells of myocardial, myometrial, epithelial and endothelial origin (reviewed in Dermietzel et aI., 1990). By contrast, connexin expression can exhibit tissue specificity. For example, connexin33 is found solely in rat testicular cells (Haefliger et al., 1992). Functions of Gap Junctions Gap junctions are involved in a number of physiological events including synchronization of pacemaker cells and propagation of electrical impulses throughout the myocardium, electrical coupling of nerve cells, embryonic differentiation and generation of rhythmic uterine contractions during parturition (reviewed in Beyer et al., 1990). In addition, gap junctional communication (gjc) is believed to regulate tissue , cell growth and proliferation by allowing the exchange of ions, small molecules such as metabolites and second messengers and possibly growth regulatory molecules having molecular masses less than 1kDa (Loewenstein and Rose, 1992). Previous studies have compared the ability of normal and transformed cells to communicate intercellularly (Mehta et al., 1986; Yamasaki, 1990). Cells transformed by the v-src or H­ ras oncogenes exhibit reduced gjc (Arzania et al., 1988; Dotto et aI., 1989; EI-Fouly et aI., 1989; Crow et al., 1990; Brisette et al., 1991). In addition, cells transformed by a temperature-sensitive mutant of Rous sarcoma virus exhibited a rapid and reversible reduction in gjc when the cells were changed from the non-permissive temperature to the permissive temperature required for cellular transformation (Atkinson et al., 1981; Azarnia and Loewenstein, 1984; Crow et aI., 1992). Growth factors and tumor promotors also disrupt gjc in a number of cell lines (Maldonado et

2 al., 1988; Madhukar et al., 1989; reviewed in Klaunig and Ruch, 1990; Berthoud et al., 1992; Lau et aI., 1992). Moreover, Mehta et al. (1991) demonstrated that expression of a transfected Cx43 cDNA in communication-deficient, transformed C3H 1OT1 /2 fibroblasts resulted in the production of functional Cx43 leading to the normalization of gjc and growth. Thus, the modulation of gjc is of particular interest because of its possible role in regulating various physiological functions. Regulation of Gap Junctional Permeability Various regulatory mechanisms control the permeability of gap junction channels. Some gap junctions are voltage-dependent. This phenomenon was originally observed in the crayfish giant motor synapse and subsequently observed in arthropod and embyonic vertebrate gap junctions as well as several mammalian gap junctions (reviewed in Dermietzel et al., 1990). In certain tissues, gap junctions are sensitive to intracellular pH to varying degrees (Spray et aI., 1985). Junctional permeability decreases as the hydrogen ion concentration increases and the complete disruption of gjc occurs rapidly and reversibly. Although the mechanism of action is as yet unclear, it has been suggested that the gap junction protein may contain specific proton binding sites such as histidine or lysine residues which may act to directly block junctional permeability (Spray et aI., 1981). Gap junctions can also be regulated by the free intracellular Ca+2 concentration where channels close in response to elevating levels of Ca+2 (Loewenstein, 1981). During cell injury, elevated levels of extracellular Ca+2 enter the cell. As a defense mechanism, gap junction channels close to prevent the loss of ions, metabolites and small molecules from neighboring cells. Several hypotheses as to the mechanism of Ca+2 action have been suggested. The first suggests that 3 the direct binding of Ca+2 to the gap junction channel blocks the passage of molecules through the channel (reviewed in Ramon and Rivera, 1986). Another suggests that Ca+2 action occurs indirectly through the interaction with the calcium binding protein, calmodulin (Peracchia,1985,1988). Whether either of these mechanisms is responsible for the Ca+2 effect remains to be determined. Recent studies have demonstrated the connexin protein to be a phosphoprotein containing phosphates on serine and/or tyrosine residues (Crowet al., 1990; Musil et al., 1990; Stagg and Fletcher, 1990; Laird et al., 1991; Oh et al., 1991; Lau et al., 1991, 1992). These studies suggest that phosphorylation may be a regulatory mechanism. However, the (s) acting directly upstream of the connexin protein remains unknown. Previous studies have demonstrated a direct correlation between increased levels of cyclic AMP (cAMP) and increased gap junction channel permeability (Flagg-Newton et al., 1981; Saez et al., 1986). These studies suggest that the cAMP-dependent protein kinase (PKA) may be involved. To further support PKA's involvement, the catalytic subunit of PKA was demonstrated to directly phosphorylate the connexin protein in vitro (Saez et al., 1986, 1990). Moreover, a synthetic peptide corresponding to the carboxy terminal domain of the connexin protein was phosphorylated by PKA producing a similar two-dimensional tryptic phosphopeptide map as that seen with the connexin protein phosphorylated in vivo (Spray et aI., 1988) In addition to the short term effects of cAMP on the permeability of gap junction channels, previous studies have suggested that cAMP may also be involved in the biosynthesis of the connexin protein (Roesler e1 aI., 1988; Risek et al., 1990). Taken together, these studies provide evidence 4 that cAMP could regulate gap junctions at the transcriptional level as well as the post-translational level. These findings are physiologically significant in that the smooth muscles of the uterus which are normally weakly interconnected via gap junctions become highly efficient in gjc prior to parturition to enable synchronized uterine contractions during birth (reviewed in Dermietzel et aI., 1990). It is believed that circulating hormones induce an increase in the levels of cAMP leading to the up­ regulation of gjc (Garfield et al., 1977; reviewed in Warner, 1988). Another intensely examined enzyme believed to act directly on the connexin protein is the Ca+2/phospholipid-dependent protein kinase (PKC). Phorbol esters such as 12-0-tetradecanoylphorbol-13-acetate (... (TPA) are potent tumor promoters that inhibit gjc in a number of cell lines (Konishi et al., 1990; Asamoto et al., 1991; Berthoud et al., 1992). To date, the molecular mechanism(s) by which these tumor promoters inhibit intercellular communication is unclear. However, it has been correlated with their ability to activate PKC (Nakamura et aI., 1989), a serine/threonine kinase described as the major phorbol ester receptor (Parker et aI., 1986; Blumberg, 1988). Previous studies have demonstrated TPA-induced phosphorylation of Cx26, Cx43 and MP26, a lens gap junction protein in vivo (Lampe and Johnson, 1989; Brissette et aI., 1991; Oh et al., 1991; Berthoud et al., 1992) or direct phosphorylation of Cx32 by purified PKC in vitro (Takedaet aI., 1987; Saez et al., 1990). In addition, diacylglycerol (DAG), the physiological activator of PKC (Bell, 1986; Ganong et al., 1986) as well as 1-0Ieoyl-2-acetyl-rac-glycerol (OAG), a synthetic DAG derivative have been demonstrated to inhibit gjc in a manner similar to that of phorbol esters (Gainer and Murray, 1985; Enomoto and Yamasaki, 1985). DAG and phorbol esters bind to and 5 activate PKC at common overlapping sites in a competitive manner. However, phorbol esters are more potent that DAG and not as rapidly metabolized (Sharkey et aI., 1984; Burns and Bell, 1991). Therefore, TPA­ induced disruption of gjc may be mediated by PKC. Oncogenic protein products have also been implicated in gap junction channel modulation. Earlier studies demonstrated that cells infected with the Rous sarcoma virus exhibited reduced junctional permeability (Atkinson et al., 1981 ; Azarnia and Loewenstein, 1984; Chang et al., 1985; Azarnia et aI., 1988) The mechanism(s) by which gjc becomes reduced is unclear. However, correlated with the reduction of gjc was the phosphorylation of the connexin protein on tyrosine residues (Crow et al., 1990, 1992; Swenson et al., 1990). In addition, cells transformed by the polyoma virus also exhibit reduced junctional permeability similar to that observed in src transformed cells (Azarnia and Loewenstein, 1987). The viral middle T antigen has been demonstrated to activate the src tyrosine protein kinase (Bolen et al., 1984). Whether the sre tyrosine kinase directly phosphorylates the connexin protein or activates a downstream tyrosine kinase is yet unclear. In addition to the src oncogene, others have reported reduced gjc in cells transformed by the ras oncogene. Total disruption of gjc, however, was induced by the synergistic effects of ras and the phosphoinositol signalling pathway activated by TPA (Dotto et aI., 1989; Paulson and Atkinson, 1989; Brissette et al., 1991). Cellular transformation by the mye, mos., fps and neu oncogenes has also been reported to reduce gjc (Atkinson and Sheridan, 1988; Bignami et al., 1988; Kurata and Lau, 1993). While these studies demonstrate that oncogenic transformation is associated with modulating

6 junctional permeability, further investigation as to the physiological significance is required. Growth factor-mediated disruption of gjc has been demonstrated in a number of cells lines (Maldonado et al., 1988; Madhukar et al., 1989). Whether this effect is regulated by phosphorylation is as yet unclear. Moreover, the cascade of events linking growth factor receptor activation to the disruption of gjc are still largely unknown. Growth Factor-Mediated Signalling Pathways Growth factors such as epidermal growth factor (EGF) and platelet­ derived growth factor (PDGF) bind to specific cell surface transmembrane receptors with intrinsic tyrosine kinase activity (reviewed in Varden and Ullrich, 1988). Upon ligand binding and oligomerization, the receptor tyrosine kinase is activated phosphorylating a number of cellular substrates on tyrosine residues as well as its own polypeptide chain (autophosphorylation). The autophosphorylation sites of the receptor are located at the carboxy terminal tail which is presumed to be the regulatory domain of the receptor (Helin et aI., 1991 ; Sorkin et al., 1991). It has been suggested that autophosphorylation may induce a conformational change of its kinase domain leading to the recruitment and/or activation of a number of substrates including Shc, Grb-2, GTPase-activating protein (GAP), Phospholipase C-gamma (PLC-gamma) and phosphoinositide 3 kinase (P13 kinase) (Ahn et ai, 1990a,b, 1991; Carpenter and Cohen, 1990; Margolis et al., 1990; Weiel et aI., 1990; Cochet et al., 1991; Hsu et al., 1991; Glenney Jr., 1992; Hernandez-Sotomayor and Carpenter, 1992; Margolis, 1992; Pelicci et al., 1992; Rozakis-Adcock et al., 1992). The recruitment mechanism for these substrates appear to utilize the src homology (SH) 2 domain/phosphotyrosine association (Anderson et aI., 7 1990; Moran et aI., 1990; Koch et aI., 1991; reviewed in Pawson and Gish, 1992). These substrates may act as intermediates regulating growth factor-induced disruption of gjc. The association of SH2-containing Shc and Grb-2 proteins to the activated receptor has been reported to lead to the activation of p21 ras (Clark et aI., 1992; Lowenstein et aI., 1992). Ras activation ultimately leads to the activation of MAP kinase (de Vries-Smits et aI., 1992; Hattori et aI., 1992; Pomerance et al., 1992; Wood et al., 1992) through PKC­ dependent and -independent signalling pathways (Vila and Weber, 1988; Nel et aI., 1990; L'Allemain et aI., 1991; Meier et at., 1992). The exact position of PKC in the Ras signalling pathway remains undefined. MAP kinase, a serine/threonine kinase, is activated in response to a variety of agents including , EGF, PDGF, and TPA (Ray and Sturgill, 1987, 1988a,b; Kazlauskas and Cooper, 1988; Rossomando et aI., 1989; Gotoh et aI., 1990a; Adams and Parker, 1991; L'Allemain et al., 1991). To date, four distinct isoforms of MAP kinase have been identified (reviewed in Pelech and Sanghera, 1992). These isoforms have been found in yeast, sea star oocytes, rats and humans, among other species (Courchesne et aI., 1989; Elion et al., 1990; Kyriakis et aI., 1991; Sanghera et aI., 1991; Gonzalez et aI., 1992). Evolutionary conservation and diverse tissue distribution suggest that MAP kinases are likely to be essential for cellular processes and may serve isoform specific functions. In response to various agonists, MAP kinase is phosphorylated on tyrosine and serine/threonine residues resulting in enzymatic activation (Ray and Sturgill, 1988b; L'Allemain et aI., 1992; Payne et aI., 1991). MAP kinases have been referred to as "switch kinases", serine/threonine protein kinases regulated by tyrosine phosphorylation, thought to be

8 involved in integrating multiple signalling pathways (Anderson et aI., 1990; Payne et al., 1991). A number of MAP kinase substrates have been identified, including ribosomal S6 kinase, EGF-receptor, c-jun, MAP-2 and transcription factor p62TCF (Ray and Sturgill, 1987; Sturgill et aI., 1988; Northwood et aI., 1991; Pulverer et aI., 1991; Takishima, et aI., 1991; Gille et al., 1992). Thus, MAP kinase may playa role in EGF's ability to disrupt gjc in T51B cells. In addition, EGF has been demonstrated to stimulate the phospholipase C (PLC)-gamma/phosphoinositol pathway producing the second messengers inositol 1,4,5-trisphosphate (IP3) and DAG which

mediate the release of intracellular Ca2+ and PKC activation, respectively (Berridge, 1987; Nishizuka, 1986,1988). It is possible that PKC may mediate growth factor-induced disruption of gjc. This signalling pathway however, is not absolutely required for cellular function and is not observed in all cell lines. Hill et al. (1990) have presented evidence that PDGF­ induced PLC-gamma activation and the subsequent accumulation of IP3

and increase in intracellular Ca2+ are not required for DNA synthesis as previously believed. Protein Kinase Phosphorylation Consensus Sequences Protein kinases recognize localized areas surrounding the residue(s) acting as the phosphoacceptor. These areas have been described as protein kinase phosphorylation consensus sequences and are believed to act as determinants for substrate specificity (Kennelly and Krebs, 1991). Although there are reports of various receptor activated kinases, PKC and MAP 2 kinase are of particular interest because

computer-assisted sequence analysis of Cx43 identified consensus sites for both kinases. PKC requires basic amino acid residues near the

9 phosphoacceptor group. R/K-X-SIT-X-R/K has been described as the consensus sequence for PKC where X can be substituted with any amino acid (Kennelly and Krebs, 1991). Three consecutive repeats of PKC consensus sequences were located at positions 361-373, near the carboxyl terminal tail of the connexin43 protein. MAP kinase requires P-L­ SIT-P as consensus sequences for substrate protein phosphorylation (Alvarez et aI., 1991). Two perfect MAP 2 kinase consensus sites were located at positions 252-255 and 276-279 of the Cx43 protein. In addition to the PKC and MAP 2 kinase consensus sites, several other sites in the connexin protein have been identified including those for p34Cdc2, glycogen synthase kinase-3 and casein kinase I (Kennelly and Krebs, 1992). Thus, it is possible that Cx43 is a target of multiple signalling pathways mediated by protein kinases. Characterization of EGF-induced Disruption of GJC In the current study, EGF's ability to disrupt gjc in T51B rat liver epithelial cells is examined. Although associated with the disruption of gjc, the significance of Cx43 phosphorylation has remained unclear. Studies in Chapter 3 examine the effects of EGF on the integrity of gap junctional plaques. In addition, the phosphorylation state of Cx43 following EGF treatment will also be examined. If Cx43 becomes phosphorylated in response to EGF, the question of whether a functional relationship exists will be addressed. Also of considerable interest is the signalling pathway(s) linking growth factor-induced receptor activation to the disruption of gjc and Cx43 phosphorylation. Studies in Chapter 4 investigate the possible signalling cascades. EGF has been demonstrated to activate the Ras signalling pathway through PKC-dependent and -independent means resulting in the activation of MAP kinase. In addition,

10 EGF has been shown to stimulate the PLC-gamma/phosphoinositol signalling pathway resulting in PKC activation. One or both of these signalling pathways may be involved in EGF's ability to modulate junctional permeability. To investigate the possible role of PKC, the effects of chronie TPA treatment on EGF's effects will be examined. Finally, if Cx43 becomes phosphorylated in response to EGF, identification of the kinase(s) acting directly upstream of Cx43 will be required. Studies in Chapter 5 examine the possible involvement of MAP kinase in mediating EGF-induced Cx43 phosphorylation. EGF's ability to activate MAP kinase in T51 B cells will be investigated. In addition, the ability of purified MAP kinase to directly phosphorylate Cx43 will be examined. Understanding the molecular mechanisms involved in regulating gap junctional channels may provide insights about the role of gap junctions in normal and pathological physiological events.

,, Figure 1 Schematic model of gap junction plaque and connexin protein. (A) Cross-section of a proposed model of a gap junction plaque. Apposed connex~ns in the lipid bilayer of the plasma membrane of adjacent cells join to form complete channels allowing the exchange of ions, metabolites and small molecules less that 1 kOa. (8) Membrane topography of the connexin protein, the functional subunit of the gap junction. The connexin protein traverses the plasma membrane four times resulting in two --' N extracellular loops, an intracellular loop and the amino and carboxy termini extending into the cytosol. The extracellular loops are believed to playa role in the "docking" process bringing two apposed connexons together. The regulatory domains modulating gap junction channel permeability are thought to reside in the cytoplasmic regions of the protein which contain protein kinase phosphorylation consensus sequences. The third membrane spanning domain contains polar amino acids and is believed to form a portion of the aqueous pore. A Gap Junction a Connexin Protein

...... w

\.. __._ ..... _----'f1/IIIfIII1"----r

Cell 2 coo-

------Chapter 2 Materials and Methods

Cell Culture T51 B rat liver epithelial cells (originally provided by Alton Boynton, Pacific Northwest Research Foundation) were cultured in BME (GIBCO) supplemented with 10% bovine calf serum (BCS; Colorado Serum Co.) and maintained at 370C in a humidified 5% C02 incubator. Northern blot analysis demonstrated that only Cx43 of the connexin cDNA's examined (Cx26, Cx32, Cx37, Cx 40, Cx43 and Cx46) was expressed in T518 cells (G. Goldberg and A. Lau, unpublished observations). In light of these results, confluent and quiescent T51B cell cultures were used for all experiments unless stated otherwise.

Assay of Gap Junctional Communication Gjc was measured by the microinjection-dye transfer method (Crow et aI., 1990). T518 cells were treated with phosphate buffered saline (PBS; control), EGF or TPA for the indicated times and concentrations. For experiments requiring chronic TPA treatment, T518 cells were preincubated with TPA (100 ng/ml) for 24 hr at 370C prior to the EGF or TPA treatments as described above. For experiments requiring okadaic acid (OA), T518 cells were treated with P8S and/or dimethyl formamide (control) or EGF (25 ng/ml) for 30 min or 3 hr in the presence or absence or OA. OA was added either 1.5 hr prior to treatment with EGF for 30 min or 1 hr following EGF addition for the 3 hr time point. Following the various treatments, individual T51 B cells were microinjected pneumatically with a glass micropipette (Flaming-Brown P80/PC micropipette puller -

14 Sutton Instrument Co.) containing 10% (w/v) Lucifer yellow dissolved in 0.33 M lithium chloride using a Zeiss micromanipulator and Eppendorf pneumatic injector. Dye transfer was visualized on a Nikon diphot phase­ contrast inverted microscope with epifluorescence. Gjc was quantitated by counting the number of cells that contained the Lucifer yellow dye one minute after microinjection of a single cell.

Immunofluorescence Microscopy T51 B cells grown to confluency on glass coverslips were treated with PBS (control) or EGF (50 ng/ml) for the indicated times. Cells were rinsed with PBS followed by fixation with 3% (w/v) paraformaldehyde in PBS for 20 min at room temperature (RT). Cells were permeabilized with 0.2% Triton X-100 in PBS for 2 min then reduced with 26 mM sodium borohydride for 15 min at RT. Following a 15 min wash in PBS, the cells were incubated with either preimmune serum or rabbit antiserum directed against a peptide encoding amino acids 368-382 of rat heart connexin43 (CT368) at a 1:1000 dilution for 2 hr at RT. Following the incubation period, the cells were washed in PBS then blocked with normal goat serum (1 :50) for 30 min at RT. The cells were incubated with fluorescein isothiocyanate (FITC)-conjugated goat anti-rabbit IgG (1 :80) for 1 hr at RT. The cells were rinsed with PBS and mounted with 50% PBS/50% glycerol containing 1 mg/ml p-phenylenediamine and visualized on a Zeiss Axioplan Universal microscope with epifluorescence.

Metabolic labeling, Immunoprecipitation and Quantitation of Cx43 T51 B cells were metabolically labeled as previously described (Crow et aI., 1990) with [32Pil (NEN, NEX-53) at 0.5 - 3 mCilml in BME-1 % BCS

15 phosphate-deficient medium (GIBCO) for 3 hr at 370C. During the labeling period, cells were treated with PBS (control), EGF or TPA for the indicated times and concentrations. Experiments requiring chronic TPA or OA treatment were conducted, as described above, prior to or during the labeling period. Following the labeling period, the cells were rinsed with PBS and lysed in RIPA buffer [150 mM NaCI, 1% sodium deoxycholate, 1% TX-100, 0.1% sodium dodecylsulfate, 10 mM Tris-HCI (pH 7.2)] and clarified by centrifugation. The Iysates were incubated with either non­ immune rabbit serum (NRS) or Cx43 CT368 peptide antiserum for 1.5 hr on ice as previously described (Crow et aI., 1990). The immunoprecipitates were incubated with protein A-S. aureus (1 :20 ratio of antisera) for 30 min on ice. The immune complexes were collected by centrifugation at 1400 X g for 15 min at 40C, washed four times in RIPA buffer and extracted by boiling in SOS sample buffer. The immunoprecipitates were analyzed by SOS-PAGE on a 7.5-15% polyacrylamide gradient gel (Laemmli et aI., 1970). Following electrophoresis, the gel was stained in Coomassie blue, destained in 30% methanol/10% acetic acid (vlv), dried and exposed to X-OMAT XAR-5 film (Eastman Kodak Co.) for 1-3 days at -700C. For two-dimensional tryptic phosphopeptide mapping experiments requiring purified Cx43, unfixed/unstained wet gels were autoradiographed overnight at 40C. Radioactive quantitation of the Cx43 proteins was accomplished by excising and rehydrating gel bands followed by scintillation counting (Beckman LS 3801 ).

16 Phosphoamino Acid Analysis T51 B cells were treated with PBS (control) or EGF (25 ng/ml) for the indicated times or preincubated with TPA (100 ng/ml) for 24 hr prior to EGF treatment as described above. Metabolic labeling, immunoprecipitation and SOS-PAGE were conducted as described above. Phosphoamino acid analysis of the phosphorylated Cx43 proteins was conducted as described by Kamps and Sefton (1989). Following SDS­ PAGE, the immunoprecipitated proteins were electrotransferred to an Immobilon P membrane (Millipore Corporation) as previously described (Matsudaira, 1987). The samples were acid hydrolyzed in 5.7 N HCI for 65 min at 11 OOC, lyophilized and mixed with a solution containing unlabeled phosphoserine, phosphothreonine and phosphotyrosine standards. The samples were spotted on 0.1 mm cellulose-coated thin layer plates and electrophoresed in a pH 1.9 buffer at 40C, 1000 V for 32 min in the first dimension then electrophoresed in a pH 3.5 buffer at 40C, 1000 V for 21 min in the second dimension. The standards were visualized by staining with ninhydrin and the plates were exposed to Kodak X-OMAT XAR-5 film with an intesifying screen (Dupont) for 21 days at -700C.

Western Blot Analysis T51B cells were preincubated with PBS or TPA (100 ng/ml) for 24 hr prior to treatment with EGF for the indicated times and concentrations. Following the treatment period, the cells were either immunoprecipitated with a polyclonal EGF receptor antibody (a gift of Dr. F. Fazioli; National Institutes of Health, Bethesda, MD) as previously described or lysed in hot SOS sample buffer, boiled for 5 min and clarified by centrifugation. The immunoprecipitates or 20 Jlg of total cellular protein was analyzed by SOS- 17 PAGE on 7.5% or 10% polyacrylamide gels and electrotransferred to an Immobilon P membrane as described earlier. The membrane containing the proteins of interest was blocked with 1% bovine serum albumin (BSA) in tris buffered saline (TBS) [10 Mm Tris (pH 7.4),0.15 M NaCI] overnight at RT. This step was conducted to prevent any nonspecific binding. Following the overnight block, the membrane was washed in TBS containing 1 mM EOTA and 0.05% Tween-20 for 30 min at RT. The membrane was incubated with either 2 Ilg/ml phosphotyrosine monoclonal antibody (Upstate Biotechnology Inc.), 21lg/ml rabbit anti-PKC zeta peptide antibody (GIBCO BRL), 1 Ilg/ml anti-pan PKC polyclonal antibody (Upstate Biotechnology Inc.) or a 1:2000 dilution of rabbit antiserum raised against the C-terminal peptide of Xenopus MAP kinase (Xp42; a gift of Drs. J. Posada and J. Cooper; Fred Hutchinson Cancer Research Center, Seattle, WA) for 2 hr at RT. The membrane was washed as described above and the immunoreactive protein bands were visualized by incubating the membrane with either [1251] goat anti-mouse IgG (1 X 106 cpm) (NEN, NEX-159)or [1251] goat anti-rabbit IgG (1 X 106 cpm) (NEN, NEX-155) for 1 hr at RT. The membrane was washed extensively as described above and autoradiographed using Kodak X-OMAT XAR-5 film with an intensifying screen for 1-3 days at -700C.

In vitro Protein Kinase Assays PKC activity was assayed as described by Meier et al. (1991). T51B cells were unstimulated or EGF-treated (25 ng/ml) for 15 min following preincubation with PBS or TPA (100 ng/ml) for 24 hr at 370C. Cells were scraped from culture dishes in extraction buffer [20 mM HEPES (pH 7.5), 10 mM EGTA, 2 mM EOTA, 2 mM OTT] and Oounce homogenized with 30

18 strokes. The crude soluble fraction was obtained after centrifugation in a TLA 100.2 rotor at 100,000 X g for 20 min at 40C in a Beckman TL-100 ultracentrifuge. The pellet was resuspended by passage through a 22G 1-1/2 inch needle in extraction buffer containing 0.1% Triton X-100 (v/v) and incubated in ice for 30 min. The crude particulate fraction was obtained following centrifugation in a TLA-100.2 rotor at 400,000 X g for 20 min at 40C in a Beckman TL-100 ultracentrifuge. PKC was partially purified by DEAE-Sephacel chromatography as described by Meier et al. (1990). All chromatography steps were conducted at 40C. The DEAE-Sephacel column (1.5 ml bed volume/ 1 ml crude extract) was equilibrated with column buffer [20 mM HEPES (pH 7.5), 2 mM EGTA, 2 mM EOTA, 2 mM OTT]. The crude cytosolic fraction was applied to the column and washed with 15 ml of column buffer. Soluble PKC activity was batch eluted in 0.1

M NaC!. The 1% Triton X-100-solubilized crude particulate fraction was applied to the column and washed as described above. Membrane-bound PKC activity was batch eluted in 0.15 M NaCI. PKC activity in OEAE­ Sephacel eluates of cytosolic and particulate fractions of T51B cells were assayed immediately following column chromatography. The 50 fll reaction mixture contained 26 mM HEPES (pH 7.5), 10 mM MgCI2, 0.6 mM EGTA, 0.6 mM EOTA, 30 mM NaCI, 5.6 mM OTT, 60 J.lg/ml phosphatidylserine (Avanti) , 50 J.lg histone Hi type IllS (Sigma), 6 J.lg/ml diolein (Sigma), 50 J.lM [gamma-32p] ATP (NEN, NEG-002Z). The assay was initiated by the addition of cytosolic or particulate fractions (2fl9) and incubated for 6 min at 300C. The reaction was terminated by the addition of 50 J.l1 2X Laemmli sample buffer and boiled for 2 min. The samples were resolved by SOS-PAGE on a 12% polyacrylamide gel as previously described. The gel was stained/destained, dried and autoradiographed as 19 previously described. Quantitation of the phosphorylated histone proteins was accomplished by excising and rehydrating the gei band followed by scintillation counting (Beckman LS 3801). Myelin basic protein (MBP) kinase activity was assayed as described by Boulton et al. (1991a). T51B cells were treated with PBS (control) or EGF for the indicated times and concentrations. For experiments requiring chronic TPA treatment, T5; B cells were preincubated with TPA (100 ng/ml) for 24 hr at 370C prior to the addition of EGF as described above. Following the treatment period, the cells were scraped from culture dishes in extraction buffer [50 mM B-glycerophosphate (pH 7.4), 1 mM EGTA, 1 mM OTT, 2 mM PMSF, 0.1 mM sodium orthovanadate] then passed through a 26G 3/8 inch needle five times. The samples were centrifuged at 14,000 rpm in a microcentrifuge for 5 min at 40C. The resulting cytosolic fraction was assayed immediately. The 30 JlI reaction mixture contained 30 mM HEPES (pH 8.0),1 JlCi [gamma-32p] ATP, 1 mM OTT, 1 mM benzamidine, 10 mM MgCI2, 100 Jlg/ml BSA, 3 Jlg MBP and 10 Jlg cytosolic fraction. The reaction was initiated by the addition of MBp/[gamma-32p] ATP or [32P-gamma] ATP alone and incubated for 10 min at 300C. The reaction was terminated by the addition of 30 JlI 2X Laemmli sample buffer and boiled for 2 min. The samples were resolved by SOS-PAGE on a 12% polyacrylamide gels, prepared for autoradiography and quantitated as previously described.

In Gel M8 P Kinase Assay MAP kinase activity was analyzed in MBP-containing polyacrylamide gels as previously described (Kameshita and Fujisawa, 1989; Tobe et aI., 1991; Wang and Erikson, 1992). Briefly, cellular extracts containing 20 Jlg

20 of protein were separated on a 10% SOS-polyacrylamide gel containing 0.4 mg/ml MBP. Following electrophoresis, the gel was washed two times with 20% isopropanol in Buffer B [50 Mm Tris (pH 8.0), 5 mM 2­ mercaptoethanol] for 1 hr at RT then reequilibrated with Buffer B alone for 1 hr at RT. The proteins were denatured with 6 M guanidine HCI in Buffer B for 1 hr at RT followed by renaturation with Buffer B containing 0.04% Tween 40 for 16 hr at 40C. Following renaturation, the gel was preincubated in kinase buffer [20 mM Tris (pH 7.2), 10 mM MgCI2, 15 mM B-glycerophosphate, 0.3 mM sodium vanadate] for 30 min a RT. The kinase reaction was initiated by incubating the gel with 5 ml kinase buffer containing 50 ~M ATP and 50 ~Ci [gamma-32p] ATP for 30 min at RT. The reaction was terminated by washing the gel extensively with 5% trichloroacetic acid (TCA, w/v) solution containing 1% sodium pyrophosphate. A control experiment performed with a gel polymerized in the absence of MBP substrate indicated that the observed MBP kinase activities were not due to autophosphorylation. The washed gel was dried and autoradiographed with Kodak X-OMAT XAR-5 film for 1-3 days at -700C. Quantitation of MBP phosphorylated by MAP kinase was measured by excising and rehydrating the gel bands and scinitillation counting as previously described.

In vitro phosphorylation of Cx43 by MAP kinase In vitro phosphorylation of Cx43 by MAP kinase was performed using the kinase buffer and assay conditions as described by Thomas et al. (1992). Cx43 was partially purified from Cx43 recombinant baculovirus­ infected Sf9 cells (Loo et aI., 1993). The cells were Dounce homogenized in homogenization buffer [50 mM Tris (pH7.5), 0.15 M NaCI, 2 mM EDTA, 21 1 mM EGTA, 25fl9/ml leupeptin, 25 uq/rn! aprotinin] with 30 strokes. The crude particulate fraction was obtained after centrifugation in a TLA 100.2 rotor at 100,000 X 9 for 20 min at 40 C in a Beckman TL-100 ultracentrifuge. The pellet was resuspended in homogenization buffer by passage through a 22 G 1-1/2 inch needle and treated at 650C for 5 min to minimize endogenous kinase activity. Heat treatment did not degrade or affect the mobility of Cx43 as determined by Western blot analysis (data not shown). 70 fl9 of the particulate fraction was incubated in kinase buffer [30 mM HEPES (pH 7.4), 10 mM MgCI2, 1 mM OTT, 20 J.lM ATP, 20 J.lCi [gamma-32p] ATP, 5 mM benzamidine] with 18 J.lg of purified MAP kinase (Ray and Sturgill, 1988a) from EL4 cells (a gift of Dr. M. Weber, University of Virginia, Charlottesville, VA) for 30 min at 300C. Kinase reactions were terminated by the addition of RIPA buffer followed by immunoprecipitation with Cx43 CT368 peptide antiserum and the immunoprecipitated proteins were resolved by SOS-PAGE , autoradiographed and quantitated as previously described.

Two-Dimensional Tryptic Phosphopeptide Mapping T51B cells metabolically labeled with [32pj] or in vitro kinase reactions as described above were solubilized in RIPA buffer, immunoprecipitated with Cx43 CT368 peptide antiserum and the proteins were resolved by SOS-PAGE as previously described. Two-dimensional tryptic phosphopeptide mapping was conducted as described by Boyle et at. (1991). The labeled Cx43 proteins were excised and eluted from unfixed, unstained wet gels in 1 ml 50 mM ammonium bicarbonate for 4 hr at RT. The eluate was precipitated with 250 III 100% TCA and 20 Ilg pancreatic RNase A as a carrier protein for 2 hr on ice. The precipitated 22 protein was collected by microcentrifugation at 12,000 rpm for 10 min at 40C. The pellet was washed in 500 J..l1 ice cold absolute ethanol and microcentrifuged as described above. The washed and dried TeA precipitate was oxidized in 50 J..l1 ice cold performic acid for 1 hr on ice and lyophilized to complete dryness. The oxidized protein was resuspended in 50 J..l1 50 mM ammonium bicarbonate and 25 J..lg 1-Tosylamide-2­ phenylethyl chloromethyl ketone (TPCK)-trypsin (Worthington) for 14 hr at 370C. 10 J..lg of TPCK-trypsin was added for an additional 2 hr at 370C. The TPCK-trysin was diluted with 500 J..l1 water and lyophilized. The phosphopeptides were washed again as above then resuspended in pH 1.9 buffer [2.5°/0 (vlv) formic acid (88%), 7.8% (vlv) glacial acetic acid] and relyophilized. The samples were resuspended in 5 J..l1 pH 1.9 buffer. The resulting tryptic peptides were separated on cellulose TLC plates (Curtin Matheson Scientific) by electrophoresis using pH 1.9 buffer for 1 hr at 1 KV in the first dimension and by ascending chromatography using isobutyric acid buffer in the second dimension. The TLC plates were exposed to Kodak X-OMAT XAR-5 film with an intensifying screen for 12-14 days at -zovc.

23 Chapter 3 EGF·induced disruption of GJC and connexin43 phosphorylation

As mentioned earlier, growth factors such as EGF have been demonstrated to reduce junctional permeability in several cell lines (Maldonado et aI., 1988; Madhukar et aI., 1989). The studies in Chapter 3 examine EGF's ability to modulate gjc in T51 B rat liver epithelial cells. The integrity of gap junctional plaques as well as the phosphorylation state of Cx43 following EGF treatment are examined.

EGF disrupts gjc in a rapid and transient manner. EGF (25 ng/ml) treatment of T51 B cells resulted in a rapid reduction of gjc where a 30% reduction (60 cells in control to 40 cells) occurred as early as 10 min following agonist addition (Figure 3.1A). Further reductions in gjc occurred at 20 min (8 cells) and reached maximum levels of reduction «2 cells) by the 1 hr time point. This EGF-induced effect was transient in that gjc was restored nearly to levels observed in control cells (50 cells) by 2 hr and maintained at that level for 24 hr (Figure 3.1 B). Control cells exhibited a slight decrease in gjc over the same time course (Figure 3.1A).

EGF-induced receptor tyrosine kinase activation. To determine the kinetics of EGF-induced receptor tyrosine kinase activation, autophosphorylation of the EGF receptor was examined. Phosphotyrosine immunoblot analysis of T51 B cells immunoprecipitated with the EGF-receptor anitbody following EGF treatment detected a protein with an apparent molecular mass of 180 kDa (Figure 3.2). The 24 immunoreactive protein most likely represents the EGF receptor. EGF stimulated a rapid increase in the phosphotyrosine content of the receptor where the earliest detectable changes occurred at 2 min and maximum levels were reached within 5 min (Figure 3.2). The phosphotyrosine content of the receptor progressively decreased nearly to levels observed in control cells by 1 hr (Figure 3.2). These findings indicate that the EGF receptor tyrosine kinase is activated well in advance of the disruption of gjc suggesting that a growth factor-induced signalling cascade may mediate EGF's effects on gjc.

EGF-induced disruption of gjc was not the result of dissociated gap junctional plaques. Treatment of cells with various agents may dissociate gap junctional plaques or induce morphological changes which may indirectly disrupt cell­ cell contacts between adjacent cells. These changes may underlie EGF's ability to disrupt gjc. Therefore, T51 B cells were analyzed by immunofluorescent microscopy using the Cx43 CT368 peptide antibody. Control and EGF treated T518 cells observed under phase-contrast appeared intact at the gross microscopic level (Figure 3.3). Distinct, punctate Cx43 antibody reactions were detected in areas of cell-cell contact (Figure 3.3). The highly reactive areas most likely represent gap junctional plaques between adjacent cells. The gap junctions appeared aggregated at localized areas throughout the periphery of the cell in control cells as well as 10 and 30 min following EGF addition when gjc was observed to be greatly diminished. The gap junctions appeared less clustered nevertheless still present between adjacent cells at the 3 hr time point, when restoration of gjc was nearly to levels observed in control cells 25 (Figure 3.3). These immunologic reactions observed in the control and EGF treated cells were specific in that the preimmune serum did not elicit similar responses for all time points examined (Figure 3.3, 30 min time point shown). Thus, at the light microscopic level, EGF's ability to disrupt gjc appears unlikely to be due to any gross alterations in gap junctional plaques.

EGF alters the migration pattern of Cx43 proteins and stimulates a rapid and transient increase in Cx43 phosphorylation. Previous studies have demonstrated post-translational phosphorylation of the connexin protein to be associated with the requlatlon of junctional permeability (Crow et al., 1990, 1992; Musil et al., 1990; Swenson et aI., 1990; Brissette et al., 1991; Berthoud et aI., 1992; Moreno et al., 1992). To determine whether the same association is observed in T51B cells, the phosphorylation state of Cx43 following EGF stimulation was examined. As shown in Figure 3.4A, Cx43 was modified by phosphorylation in response to EGF. In mock-treated cells, the 45 kDa (Cx431) species was predominant. A minor 47 kDa (Cx432) band was also present (Figure 3.4A, lane 2). Within 10 and 30 min of EGF addition, the phosphorylation state of Cx43 markedly increased 2.1- and 2.9-fold, respectively, resulting in a predominant Cx432 species in addition to the appearance of a novel >47kDa (Cx433) species (Figure 3.4A, lanes 4 and 6). These marked increases in Cx43 phosphorylation occurred concomitantly with decreased junctional permeability. Following 3 hr EGF treatment, when gjc was restored nearly to levels observed in control cells, Cx43 phosphorylation levels also returned nearly to levels observed in control cells (Figure 3.4A, lane 8). These results are in agreement with

26 previous studies demonstrating the association between junctional permeability and the phosphorylation states of the connexin protein.

EGF stimulates the phosphorylation of Cx43 on serine residues. To determine which amino acid residues are phosphorylated in response to EGF, phosphoamino acid analysis as described by Kamps and Sefton (1989) was conducted on the phosphorylated Cx43 proteins. Phosphoamino acid analysis revealed that all Cx43 species contained only phosphoserine residues for all time points examined (Figure 3.48). These results suggest that a serine kinase(s) is acting directly upstream of Cx43 in response to EGF stimulation.

EGF-induced disruption of gjc and Cx43 phosphorylation are dose dependent. EGF's ability to disrupt gjc was dose-dependent where the lowest tested concentration of 0.5 ng/ml resulted in a 60% reduction (from 68 cells to 28 cells) in gjc (Figure 3.5). Further reductions in gjc occurred for up to a concentration 25 ng/ml. EGF-induced Cx43 phosphorylation was also dose-dependent. At a concentration of 0.5 ng/ml, Cx43 phosphorylation levels increased approximately two-fold and continued to increase for up to a concentration 10 ng/ml (Figure 3.5) A slight decrease in Cx43 phosphorylation was observed at a concentration 25 ng/mJ. These changes in Cx43 phosphorylation were inversely correlated with those observed in gjc which suggested that a functional relationship may exist.

27 Okadaic acid prevents the dephosphorylation of Cx43 and the restoration of gjc.

_ •• 4 To further investigate the possible functional relationship between Cx43 phosphorylation and disruption of gjc, the effects of okadaic acid (OA), a serine threonine phosphatase inhibitor (Bailojan and Takai, 1988; Haystead et aI., 1989) were examined. If the disruption of gjc is dependent on Cx43 phosphorylation, preventing the dephosphorylation of Cx43 should prevent the restoration of gjc. OA treatment for 2 hr during the 3hr EGF treatment period prevented the restoration of gjc usually observed following 3 hr EGF treatment (Figure 3.6). Occurring concomitantly was OA's ability to prevent the dephosphorylation of Cx43 (Figure 3.6, lane 12). OA also reduced gjc in the 30 min EGF-treated cells beyond the levels of disruption generally observed (Figure 3.6). Furthermore, the basal levels of gjc observed in control cells were also reduced. Consistent with the 3 hr time point, hyperphosphorylation of the connexin protein was also observed in control as well as 30 min EGF-treated cells (Figure 3.6, lanes 4 and 8). These data strongly suggest that EGF-induced disruption of gjc is functionally dependent on Cx43 phosphorylation.

28 80 A B e-e CONTROL 0--0 25 ng EGF 80 EGF(25 ng)

60 ~,. T ~O 60 1 rz:l~ ~t' 1 "". ___41, ~~ o .1\. 09900666° 0 u;;: \'-.-. ~ L u~ 09 L L - .. ~u 0 .... rr..u 40 40 0 .... ll:Z T ~Z rz:l::> 0 ~::> 1II::a 1Il::!l ::a::a :::a::!l 20 ::>0 ::>0 Zu 20 Zu 0 0 \O. T .._0-0 <, 0 • L "(.) o 10 20 30 40 50 60 0 4 8 12 16 20 24 TIME (min) TIME (hr)

Figure 3.1 EGF-induced disruption of gjc in T51B rat liver epithelial cells. T51 B cells were treated with EGF (25 ng/ml) for up to 60 min (A) or 24 hr (8) and assayed for gjc at the indicated times as described in the Materials and Methods. Mock-treated cells were exposed to PBS for the indicated times prior to assaying gjc. Each point represents the average (± SEM) of 15-20 determinations. 29 c - leOK "

Antiserum: NI---E NI-- E Nt E NI E NI E Nt E Time (min): Mock 2 5 10 30 60

Figure 3.2 Time-dependence of EGF receptor activation. T51 B cells were treated with EGF (50 ng/ml) for the indicated times, lysed in RIPA buffer and immunoprecipitated with either nonimmune serum (NI) or EGF receptor poly clonal antiserum (E) as described in the Materials and Methods. The immunoprecipitated proteins were resolved by SDS-PAGE and electrotransferred to an Immobilon P membrane. The activated EGF receptor was detected by immunoblotting with phosphotyrosine antibody and autoradiographed using Kodak X-OMAT XAR-5 film with an intensifying screen for 5-7 days at -70oe. These results are representative of 2 individual experiments. 30 Figure 3.3 Immunofluorescent localization of Cx43 in mock- and EGF­ treated T51B rat liver epithelial cells. T51B cells were treated with PBS (control) or EGF (25 ng/ml) for the indicated times. The cells were fixed, permeabilized and incubated with either preimmune serum or Cx43 CT368 peptide antiserum as described in the Materials and Methods. The immunoreactive complexes were visualized by a secondary reaction with fluorescein isothiocyanate (FITC)-conjugated antiserum. Fluorescent and phase-contrast images were examined and photographed on a Zeiss Axioplan Universal microscope with epifluorescence. Results shown are representative of 2 individual experiments.

31 Fluorescent Phase

Mock

Anti­ Cx43

30"

3hr

preimmune serum

32 Figure 3.4 EGF-induced serine phosphorylation of Cx43. (A) T51B rat liver epithelial cells were metabolically labeled with [32p] orthophosphate. During the labeling period, the cells were treated with either PBS (mock) for 30 min or EGF (25 ng/ml) for the indicated times. The cells were lysed in RIPA buffer and immunoprecipitated with either nonimmune rabbit serum (lanes 1,3,5,7 and 9) or Cx43 CT368 peptide antiserum (lanes 2,4,6,8 and 10). The immunoprecipitated proteins were resolved by SDS­ PAGE and autoradiographed as described in the Materials and Methods. w w (B) Phosphoamino acid analysis of Cx432 from EGF-treated T51B cells. The samples were acid hydrolyzed in 5.7 N Hel, lyophilized and mixed with standards as described in the Materi~ls and Methods. The samples were eletrophoresed using pH 1.9 buffer in the first dimension and pH 3.5 buffer in the second dimension as indicated at the botton right corner. The positions of unlabeled phosphoserine (P-S), phosphothreonine (P-T) and phosphotyrosine (P-Y) standards are outlined. Results shown are representative of 2-4 individual experiments. .'. ..b .. ' 2 . .' .,. .0 .:.

. •...... 4Ii' . o.,.

' ......

~ >, ~ (/). ~ d." .. 0 • :t

34 80 5

Z 4 (J) 60 zQ ..JCJ -I- ..JZ We::( W- W...J Of- :1: :::»0 j::D. Z

0 0 0 5 1 0 1 5 20 25 [EGF] ng/ml

Figure 3.5 Dose-dependence of EGF-induced disruption of gjc and Cx43 phosphorylation. T51 B cells were treated with EGF (30 min) for the indicated concentrations and either assayed for gjc as described in the Materials and Methods following the treatment period or 32P-labeled during the treatment period. Cells were lysed-in RIPA buffer and Cx43 was immunoprecipitated from the lysate with Cx43 CT368 peptide antiserum as described in the Materials and Methods. The immunoprecipitated proteins were resolved by SOS-PAGE and autoradiographed with Kodak O-MAT XAR-5 film for 1-2 days at -20oC. The phosphorylated Cx43 proteins were excised from the gel, rehydrated and quantitated by scintillation counting. These data represent the average (± SEM) of 2-4 independent experiments. 35 Figure 3.6 Effects of okadalc acid on EGF-induced modulation of gjc and Cx43 phosphorylation. T51B cells were treated with PBS (control) or EGF (25 ng/ml) for 30 min or 3 hr in the presence or absence of okadaic acid (1 nM). Gjc was assayed by the dye-transfer method as described in Materials and Methods. The top panel represents an autoradiogram of [32p] Cx43 from T51B cells treated as follows: lanes 1 and 2, control; lanes 3 and 4, okadaic acid; lanes 5 and 6, EGF for 30 min; lanes 7 and 8, EGF for 30 min plus okadaic acid; lanes 9 and 10, EGF for 3 hr; lanes 11 and 12, EGF for 3 hr plus okadaic acid. The CX432,3 represents Cx432 and Cx433 proteins that have not resolved well due to interference by the immunoglobin heavy chain migrating immediately above Cx433. [32p]_ labeled companion cells were lysed in RIPA buffer and the lysate was immunoprecipitated with either nonimmune rabbit antiserum or Cx43 CT368 peptide antiserum. The immunoprecipitated proteins were resolved by SOS-PAGE and autoradiographed using Kodak X-OMAT XAR-5 film for 1-3 days at -200C. The phosphorylated Cx43 proteins were excised from the gel, rehydrated and quantitated by scintillation counting. Results shown are representative of 3 independent experiments.

36 1 2 3 4 5 6 7 8 9 10 11 12

"~"'4~ ; ,. II 80 " " "

CTftL;OA

37 Chapter 4 EGF-induced disruption of gjc and connexin43 phosphorylation may be mediated through PKC-indpendent signalling pathways.

The studies in Chapter 3 demonstrated EGF's ability to disrupt gjc and phosphorylate Cx43 on serine residues in T51B rat liver epithelial cells. However, the intermediate events mediating EGF receptor activation to modulating junctional permeability and Cx43 phosphorylation have not been determined. The studies in Chapter 4 examine the possible role of PKC in mediating these EGF-induced events.

Down-modulation of TPA-sensitive PKC does not affect EGF-induced disruption of gjc. EGF (25ng/ml) or TPA (10 ng/ml) treatment of T51B cells resulted in rapid reductions in gjc (Figure 4.1 A). A 60% reduction in gjc occurred 10 min following EGF addition which reached maximum levels of disruption (76% reduction) by 30 min (Figure 4.1A, open squares). Gradual reestablishment of gjc occurred at the 1 hr time point with restoration nearly to levels observed in control cells occurring by 3 hr. These results are essentially similar to those reported previously (Lau et aI., 1992). Gjc was reduced by 47% and 79% following TPA treatment for 10 and 30 min, respectively (Figure 4.1 A, open circles). The lowest levels of gjc (99% reduction) were observed at the 1 hr time point and unlike the restoration in gjc observed in EGF treated cells, disruption of gjc persisted at 3 hr following TPA addition. The time course of TPA-induced disruption of gjc is similar to that observed in IAR 20 rat liver epithelial cells (Asamoto et aI., 1991). Preincubation of T51 B cells with TPA (100 ng/ml) for 24 hr resulted 38 in the desensitization of cells to subsequent TPA treatments where only nominal decreases in gjc were observed 10 and 30 min following acute TPA treatment (11% and 5% reduction, respectively) (Figure 4.1A, closed circles). Gjc appeared slightly elevated above control levels at the 1 hr and 3 hr time points. Conversely, chronic TPA exposure of T51 B cells did not affect EGF's ability to alter gjc where the time course and levels of disruption and subsequent restoration of gjc closely paralleled that observed in EGF-treated naive cells (Figure 4.1 A, compare solid and open squares). Thus, EGF's ability to disrupt gjc does not appear to require TPA-sensitive PKC.

Down-modulation of TPA-sensitive PKC does not affect EGF-induced Cx43 phosphorylation. To determine if the observed changes in gjc caused by EGF and TPA correlated with phosphorylation of Cx43, [32P]-labeled cell Iysates were immunoprecipitated with Cx43 CT368 peptide antiserum and the proteins were resolved by SOS-PAGE. A 2.1-fold increase in Cx43 phosphorylation occurred as early as 10 min following EGF addition with maximum levels of phosphorylation (2.9-fold increase) occurring at the 30 min time point (Figure 4.1 B, open squares). These changes in Cx43 phosphorylation inversely correlated with those observed in gjc. The levels of Cx43 phosphorylation decreased at the 1 hr time point, approaching levels observed in control cells by 3 hr following EGF addition. TPA treatment also resulted in 2.0- and 2.7-fold increases in Cx43 phosphorylation at 10 and 30 min, respectively (Figure 4.1 B, open circles). In contrast to decreasing levels of Cx43 phosphorylation observed 1hr and 3 hr following EGF treatment, maximum levels (3.2-fold increase) of Cx43 39 phosphorylation occurred 1 hr following TPA treatment. Moreover, Cx43 phosphorylation levels remained elevated at the 3 hr time point which correlated with the prolonged disruption of gjc (Figure 4.1 A). Acute TPA treatment of T51 B cells following chronic TPA exposure resulted in little change in Cx43 phosphorylation like that observed in control cells (Figure 4.1 B, closed circles). However, chronic TPA exposure did not affect EGF's ability to induce Cx43 phosphorylation in that the time course and levels of Cx43 phosphorylation and dephosphorylation closely paralleled those observed for EGF treatment of naive cells (Figure 4.1 B, closed squares). These latter results were most significant since they suggested that EGF's action on Cx43 phosphorylation was independent of TPA­ sensitive PKC activity.

Chronic lPA treatment of 151B cells down-modulates protein kinase C activity. In order to confirm that 24 hr TPA treatment of T51 B cells down­ regulated PKC activity, DEAE-Sephacel eluates of cytosolic and particulate fractions of T51B cells were assayed for PKC activity using histone type IllS as a substrate, in the absence or presence of Ca2 +, phosphatidylserine (PS) and diacylglycerol (DAG). The PKC activity values shown in Figure 4.2 represent the relative increases in histone phosphorylation over the nominal activity observed in the absence of Ca2+, PS and DAG. PKC activity was predominantly localized in the cytosolic fraction (6.3-fold increase) of unstimulated T51 B cells with the membrane-bound PKC showing limited activity (2.1-fold increase) (Figure 4.2). However, upon EGF treatment for 15 min, membrane-bound PKC activity markedly increased 6.3-fold above the nominal kinase activity 40 (Figure 4.2). Increased PKC activity in the particulate fraction of T51B cells in response to EGF was presumably due to PKC translocation to the membrane following activation (reviewed in Niedel and Blackshear, 1986). PKC activity in the cytosolic fraction of EGF-treated cells was slightly reduced from that observed in unstimulated cells (Figure 4.2). Most significantly, chronic TPA treatment greatly diminished PKC activity in both cytosolic and particulate fractions of T51B cells (Figure 4.2). In addition, EGF no longer stimulated PKC activity in cells chronically treated with TPA (Figure 4.2). These findings strongly suggested that 24 hr TPA treatment of T51 B cells results in the down-modulation of TPA-sensitive PKC.

EGF-induced connexin43 phosphorylation following chronic TPA treatment occurs on serines. As shown previously, chronic TPA treatment did not affect EGF's ability to stimulate the phosphorylation of Cx43 (Figure 4.1B) and the kinetics of Cx43 phosphorylation and dephosphorylation appeared similar to that of EGF alone. Phosphoamino acid analysis demonstrated that Cx43 was phosphorylated only on serine residues in response to EGF following chronic TPA treatment (Figure 4.3). These profiles were maintained for all Cx43 species at each time point examined (Figure 4.3, Cx432 species shown). These results indicate that chronic TPA treatment of T51 B cells does not induce the activation of a threonine or tyrosine kinase which directly phosphorylates Cx43.

41 EGF stimulates the phosphorylation of Cx43 at numerous sites in vivo. The phosphorylation sites of 32P-labeled Cx43 from T51B cells were examined by two-dimensional tryptic phosphopeptide mapping as previously described (Boyle et aI., 1991). Tryptic digestion of Cx43 from untreated T518 cells generated numerous phosphopeptides which were nominally phosphorylated (Figure 4.4A). However, EGF treatment for 30 min resulted in marked increases in phosphorylation of the phosphopeptides (Figure 4.48). The tryptic phosphopeptide map of Cx43 from cells chronically treated with TPA appeared similar to that observed for untreated cells although several phosphopeptides showed slightly higher levels of phosphorylation (compare Figures 4.4A and 4.4C, open arrows). EGF treatment following chronic TPA treatment resulted in a tryptic phosphopeptide map very similar to that of EGF alone with the exception of several missing phosphopeptides, possibly containing the TPA-sensitive PKC phosphorylation sites (Figure 4.48, closed arrows). These results indicated that Cx43 becomes highly phosphorylated on numerous sites in response to EGF. Furthermore, chronic TPA treatment does not affect EGF's ability to stimulate the phosphorylation of these sites with exception of three phosphopeptides which may contain TPA-sensitive PKC phosphorylation sites. Thus, EGF appears to stimulate the phosphorylation of Cx43 on numerous sites, possibly mediated by the actions of several serine protein kinases.

42 Figure 4.1 Effect of chronic TPA treatment on EGF-induced disruption of gjc and Cx43 phosphorylation in T51B rat liver epithelial cells. (A) T51B cells were preincubated with PBS or TPA (100 ng/ml) for 24 hr prior to acute treatment with EGF (25 ng/ml) or TPA (10 ng/ml) and assayed for gjc at the indicated times as described in the Materials and Methods. These data represent the averages (± SEM) of 8-17 determinations. EGF­ and TPA-induced disruption of gjc was determined relative to the average control value of 101 cells communicating in the absence of 24 hr TPA treatment and 72 cells communicating in the presence of 24 hr TPA treatment. (8) T51B cells treated as described above were metabolically labeled with 32Pi and Cx43 was immunoprecipitated from cell Iysates with Cx43 CT368 peptide antiserum as described in the Materials and Methods. The proteins were resolved by 50S-PAGE and autoradiographed using Kodak X-OMAT XAR-5 film for 1-3 days at -20oC. The Cx43 proteins were excised from the gel, rehydrated and quantitated by scintillation counting. Increases in Cx43 phosphorylation were determined relative to the average control value of 197-248 cpm. These data represent the averages of (+ SEM) of 2-3 individual experiments.

43 140

I: A 0 :;: 120 co .2 I: :::J_ 100 Ea E':= 01: 80 0 0_0 ~# 60 0- :;: 0 I: 40 --a- EGF :::J -

4 B --a- EGF -0- TPA 24hTPA+EGF 24hTPA+TPA CC -.- _0 3 (1)= -- en.!! w>- (I)".. 0 O.c -enCc. 2 (1)0 ~.c Gin. -CO) c::)(Q)l\:t 0 1

O-t--.--.-----.---r--.--.---,----,..----,---i o 20 40 60 80 100 120 140 160 180 200 Time (min)

44 Figure 4.2 Effect of chronic TPA treatment on EGF-induced PKC activation in cytosolic and particulate fractions of 151 B cells. lSi B cells were preincubated with PBS or TPA (100 ng/ml) for 24 hr prior to treatment with PBS or EGF (25 ng/ml) for 15 min. Crude cytosolic and particulate fractions were fractionated on DEAE-Sephacel as described in the Materials and Methods. Cytosolic and membrane-bound PKC was batch­ eluted with O.1M and 0.15M NaCI, respectively. DEAE-Sephacel eluates (2 J.lg) of cytosoJic and particulate fractions were assayed for PKC activity in the presence or absence of Ca2+, phosphatidylserine and diacylglycerol. The samples were resolved by SDS-PAGE and autoradiographed using Kodak X-OMAT XAR-5 film for 1 day at -200C . The phosphorylated histone proteins were excised from the gel, rehydrated and quantitated by scintillation counting. The nominal histone phosphorylation observed in the absence of Ca2+, phosphatidylserine and diacylglycerol was regarded as nonspecific kinase activity and the values represented in Figure 4.2 are expressed relative to this nonspecific kinase activity. Each point represents the average (+ SEM) of two independent experiments.

45 8 lZJ Cytosol 7 ffd Particulate .-~ .~- 6 0

0 Ctrl 1S'EGF 24h TPA 24h TPA+EGF Treatments

46 ...... ' .. .. ,- .. ' Il .. '.P-S ~ .' ",'II ••• .. ", ,:...... ·O.IT" • ., : ·P-T •. "...... " -...... " ...... o .. ° :p-y ...." o.- .. 0, ­.. .. 0 ...... 0 .. 0 . ".... •••• e • Mock 10' 30' 180'

Figure 4.3 Phosphoamino acid analysis of Cx432 from EGF-stimulated T51B cells chronically treated with TPA. T51B cells were preincubated with TPA (100 ng/ml) for 24 hr prior to treatment with EGF (25 ng/ml) for the indicated times. The samples were prepared, electrophoresed, electrotransferred and autoradiographed as described in the Materials and Methods and Figure 3.4. The positions of unlabeled phosphoserine (P-S), phosphothreonine (P-T) and phosphotyrosine (P-Y) standards are outlined. The directions of first and second dimension electropheresis are indicated at the bottom right corner. 47 Figure 4.4 Tryptic phosphopeptide maps of 32P-labeled Cx43. Cx43 from 32P-labeled T51B cells either untreated (A), EGF-treated for 30 min (8), TPA-treated for 24 hr (C) and EGF-treated for 30 min following chronic TPA treatment (0) were immunoprecipitated with Cx43 CT368 peptide antiserum, resolved by SDS-PAGE and autoradiographed. Cx43 was ..p.. co eluted from unfixed, unstained wet gels, digested with TPCK-trypsin and analyzed by electrophoresis in the first dimension (pH 1.9 buffer) and ascending chromatography (isobutyric acid butter) in the second dimension as described in the Materials and Methods. The closed arrows-indicate the positions of three phosphopeptides possibly containing TPA-sensitive PKC phosphorylation sites. The origin is indicated by an arrowhead. 1 :'1: -. .,;;ii\\t,:''',".,il',"',:\':,.,

, ,

· .., ••.,," .

Q

u

49 Chapter 5 EGF-induced Cx43 phosphorylation may be mediated by MAP kinase.

The results in Chapter 4 suggest that EGF is able to stimulate the disruption of gjc and Cx43 phosphorylation independent of TPA-sensitive PKC activity. EGF's ability to activate MAP kinase independent of PKC activity, taken together with the presence of two MAP kinase phosphorylation consensus sites in Cx43, suggest that MAP kinase may mediate EGF's effects. Chapter 5 examines the effects of EGF on MAP kinase activity in T51B rat liver epithelial cells as well as the possible involvement of MAP kinase in Cx43 phosphorylation.

EGF alters the migration pattern of MAP kinase isoforms in a time­ and concentration-dependent manner. EGF stimulated a rapid shift in mobility of p42 and p44 isoforms of MAP kinase to slower migrating species as determined by immunoblotting analysis using rabbit antiserum against MAP kinase (Figure 5.1 A). These changes occurred as early as 2 min following EGF addition (the earliest time point examined). The slower migrating, apparently higher molecular weight forms which predominated at the 2, 5 and 10 min time points progressively reverted to the faster migrating forms observed in control cells by 1 hr (Figure 5.1A). EGF's ability to modify the migration pattern of MAP kinase was also dose-dependent (Figure 5.1 B) with effects observed at a concentration of 0.5 ng/ml (the lowest tested concentration) which was also demonstrated to be effective at diminishing gjc (Lau et aI., 1992). The altered migration pattern was observed for up to a concentration of 25 ng/ml (Figure 5.1 B). Chronic TPA treatment had no affect on the time 50 course or dose-dependence of EGF-induced changes in MAP kinase mobility (Figures 5.1 A and 5.1 B).

EGF stimulates the tyrosine phosphorylation of MAP kinase. We were interested in determining if the observed shifts in MAP kinase mobility were associated with changes in its phosphorylation on tyrosine which has been correlated with its enzymatic activation (Ray and Sturgill, 1988b; Rossomando et aI., 1989). EGF treatment of T51 B cells resulted in a rapid tyrosine phosphorylation of both p42 and p44 isoforms of MAP kinase as determined by immunoblotting analysis utilizing phosphotyrosine (Ptyr) monoclonal antibody (Figure 5.1A). Tyrosine phosphorylation of both isoforms occurred as early as 2 min following EGF addition and reached maximum levels at 5 min, then progressively decreased nearly to control cell levels by 1 hr (Figure 5.1A). Only the slower migrating species of each of the p42 and p44 MAP kinase isoforms appeared to contain Ptyr. This EGF-induced effect was also dose­ dependent where the lowest tested EGF concentration of 0.5 ng/ml resulted in a marked increase in Ptyr content of both isoforms of MAP kinase (Figure 5.1B). This level of tyrosine phosphorylation was maintained for up to a concentration of 25 ng/ml (Figure 5.1 B). Chronic TPA treatment did not affect the time course or dose-dependence of EGF's ability to stimulate Ptyr accumulation in both p42 and p44 isoforms of MAP kinase (Figures 5.1 A and 5.1 B). Thus, it appears that EGF stimulated the tyrosine phosphorylation of p42 and p44 isoforms of MAP kinase which suggest that both may have been activated.

51 EGF stimulates activation of the p42 and p44 isoforms of MAP kinase. To measure EGF activation of MAP kinase directly, cytosolic fractions of EGF-stimulated T51B cells were assayed for myelin basic protein (MBP) kinase activity. We found that phosphorylation of MBP by EGF-activated MBP kinase occurred in a rapid and transient manner in vitro. A 2.9-fold increase in MBP kinase activity was observed 2 min following EGF stimulation which reached maximum levels within 5 min (4.2-fold increase) and progressively declined nearly to the control cell levels by 1 hr (Figure 5.2A, open squares). Preincubation of T51B cells with TPA (100 ng/ml) for 24 hr did not interfere with EGF's ability to stimulate MBP kinase activity since the time course of its activation closely paralleled that of EGF alone (Figure 5.2A, closed squares). EGF-induced MBP kinase activation was also dose-dependent. Treatment of T518 cells with increasing concentrations (0.5-25 ng/ml) of EGF tor 5 min resulted in a marked increase in M8P kinase activity which appeared to plateau at 10 ng/ml (Figure 5.28, open squares). A concentration of 0.2 ng/ml was required to reach half maximal stimulation of MBP kinase activity (Figure 5.28). A very similar dose-response curve was observed for cells chronically treated with TPA prior to stimulation by EGF (Figure 5.2B, closed squares). The cytosolic fraction of T51 B cells was assayed in situ to confirm that the MBP kinase activity was actually due to MAP kinase. Cytosolic fractions were resolved by SOS-PAGE in MBP-containing polyacrylamide gels and the kinase reaction was conducted directly in the gel containing the renatured proteins. We found that EGF transiently stimulated 44 kDa and to a lesser extent, 42 kDa MBP kinase activities which reached maximum levels within 5 min of EGF treatment (Figure 5.2C). This activity 52 progressively decreased nearly to levels observed in control cells by 60 min (Figure 5.2C). No other protein kinases which phosphorylated MBP were detectable throughout the gel. Control experiments conducted using a gel polymerized in the absence of MBP substrate, indicated that the observed MBP kinase activities were not due to autophosphorylation (negative control data not shown). These combined results strongly suggested that the cytosolic fractions of EGF-treated T51B cells contain p42 and p44 MBP kinases which electrophoretically and kinetically behave as activated MAP kinases. Furthermore, EGF-activation of MAP kinase persists in T51 B cells chronically treated with TPA and correlates with the phosphorylation of Cx43 and disruption of gjc.

MAP kinase phosphorylates Cx43 in vitro. To determine if Cx43 is a direct substrate of MAP kinase, partially purified Cx43 in particulate fractions of Cx43 recombinant baculovirus­ infected 8f9 cells were subjected to phosphorylation by purified MAP kinase in vitro (a gift of Dr. M. Weber). We discovered that phosphorylation of Cx43 in the presence of MAP kinase was 5.1-fold (± 1.1 8EM) above that observed in the absence of MAP kinase (Figure 5.3). In vitro phosphorylated Cx43 contained only phophoserine as determined by phosphoamino acid analysis (L. Loo and A. Lau, unpublished results). The slight phosphorylation of Cx43 observed in the absence of MAP kinase was possibly due to endogenous insect cell kinases which survive the heat inactivation process designed to reduce endogenous kinase activity present in 8f9 fractions (M. Kanemitsu and A. Lau, unpublished results). Nonetheless, in the presence of MAP kinase, Cx43

53 was markedly phosphorylated in vitro suggesting that Cx43 may be an in vivo substrate of MAP kinase.

Common phosphopeptides exist in Cx43 phosphorylated in vivo and by MAP kinase in vitro. Two-dimensional tryptic phosphopeptide mapping of Cx43 phosphorylated by MAP kinase in vitro revealed that several unique sites were phosphorylated directly by MAP kinase (Figure 5.4A, phosphopeptides a-d). One phosphotryptic peptide appeared to represent the major site of in vitro phosphorylation (Figure 5.4A, phosphopeptide b). Whether these phosphopeptides contain unique phosphorylation sites or are the result of partial tryptic digestion is not yet clear. The endogenous 8f9 insect cell kinase activity described earlier may have also contributed to the phosphorylation of these minor sites. Most significantly, these in vitro produced phosphopeptides appeared to migrate very similarly to phosphopeptides produced from Cx43 labeled in vivo after EGF-treatment (compare Figures 5.4A and 5.48). Comigration experiments confirmed that the in vitro produced phosphotryptic peptides did indeed represent a unique subset of the in vivo labeled ones (compare Figure 5.4C with Figures 5.4A and 5.48). These results suggested that MAP kinase may phosphorylate Cx43 directly in vivo.

54 Figure 5.1 Time- and dose-dependence of EGF-induced mobility shift and tyrosine phosphorylation of p42 and p44 lsotorms of MAP kinase. T51B cells were untreated or EGF-treated for the indicated times (A) and concentrations (8) following preincubation in PBS or TPA (100 ng/ml) for 24 hr. The cells were fractionated, resolved by SOS-PAGE and electrotransferred to an Immobilon P membrane as described in the Materials and Methods. The proteins were immunoblotted with either rabbit antiserum against the C-terminal peptide of Xenopus MAP kinase or phosphotyrosine monoclonal antibody and the immune complexes were detected with [1251] goat anti-rabbit IgG or [1251] sheep anti-mouse IgG, respectively. The membrane was autoradiographed using Kodak X-OMAT XAR-5 film for 1-3 days with an intesifying screen at -700C. The p42 and p44 isoforms of MAP kinase are marked. Results shown are representative of 2-4 experiments.

55 A a MAP Kinase ex PTyr

-p44- .~:~~-. -p42- -_._- _ ......

-p44­ +24hTPA ~. -10.1-...42 - .;;

025103060 o 2 5 10 30 60 Time(min)

B a MAP Kinase ~'.' .~~~TYr ...... •" . -p44­ --~ , 4 __ .. .-p42- --

-p44- +24hTPA -p42-

o 0.5 1.0 2.5 10 25 0 0.5 1.0 2.5 10 25 EGFConcentration (ng/ml)

56 Figure 5.2 Time- and dose-dependence of EGF-induced MBP kinase activation. T51B cells were treated with EGF for the indicated times (A) and concentrations (8) following preincubation with PBS (open squares) or TPA (100 ng/ml, closed squares) for 24 hr. Cytosolic fractions (10 Jlg) were assayed for MBP kinase activity in vitro as described in the Materials and Methods. The samples were resolved by SOS-PAGE and autoradiographed. Phosphorylated MBP was excised from the gel, rehydrated and quantitated by scintillation counting. These data represent the averages (± SEM) of 2 individual experiments. (C) Cytosolic fractions (20 Jlg) of T51B cells were resolved by 80S-PAGE in MBP-containing gels followed by denaturation with 6 M guanidine Hel and renaturation, and the kinase assay was performed in situ as described in the Materials and Methods. The arrows indicate p42 and p44 isoforms of MAP kinase.

57 A 6.0 -0- EGl' -m- 24hTPA+EGF

.5~ 5.0 Gl> ~ti::::- 4.0 Gl< 0 l:iGl~ .5~o 3.0 Gl,5° ~~~ elSa. 2.0 'Q)1Il a::l: 1.0

0.0 0 10 20 30 40 50 60 Time (min) B

4.0

-0- EGF -+- 24hTPA+EGF 0.0 0 5 10 15 20 25 EGF ~oncentration (ngfml)

C 180- 116 - 84- 58- 48.5- .... p44 "--GIIlID....., .... p42 36.5- - 26.6- kDa 0 2 5 10 30 60 Time(min)

58 ·68­ 60-

Figure 5.3 Phosphorylation of Cx43 by MAP kinase in vitro. Cx43 from particulate fractions (70~g) of Cx43 recombinant baculovirus-infected 8f9 cells were incubated in the absence (-) or presence (+) of purified MAP kinase (18 ~g) as described in the Materials and Methods. Cx43 was immunoprecipitated with the Cx43 CT368 peptide antiserum, resolved by 80S-PAGE and autoradiographed using Kodak X-OMAT XAR-5 film for 1­ 2 days at -20oC. 59 Figure 5.4 Tryptic phosphopeptide maps of Cx43 metabolically labeled in vivo and phosphorylated by purified MAP kinase in vitro. Cx43 from particulate fractions of Cx43 recombinant baculovirus-infected Sf9 cells and phosphorylated by purified MAP kinase in vitro (A) and 32P-labeled T51B cells treated with EGF for 30 min following chronic TPA treatment (B) were immunoprecipitated with Cx43 CT368 peptide antiserum and en resolved by SOS-PAGE as described in the Materials and Methods. o Two-dimensional phosphotryptic peptide analysis of Cx43 was performed as described in the Materials and Methods and Figure 4.4. Phosphopeptides representing possible MAP kinase phosphorylation sites are designated a-d. In mixing experiments (C), comigrating phosphopeptides are designated a-d. The origin is indicated by an arrowhead. ., u .i..~

•••

u

•.a

61 Chapter 6 Discussion

Intercellular communication is responsible for the coordination and regulation of cellular signals in tissue and organ systems. Ubiquitous expression and evolutionary conservation suggest that gap junctions are fundamental components of physiological processes. While the functional significance of gjc is becoming clearer, the regulatory mechanisms involved are as yet undefined. The current study attempts to gain a better understanding of the manner in which gap junctions are modulated, specifically by phosphorylation/dephosphorylation mechanisms. In addition, the signalling cascade(s) as well as the enzyme(s) linking growth factor receptor tyrosines to gap junctional modulation are examined. In agreement with previous studies (Maldonado et aI., 1988; Madhukar et aI., 1989) we have demonstrated EGF-induced disruption of gjc in T51 B rat liver epithelial cells (Figure 3.1, Lau et aI., 1992). EGF stimulated a rapid and transient disruption of gjc in T518 cells where the kinetics of disruption was similar to that observed by Maldonado et al. (1988) in normal rat kidney (NRK) or Balb C 3T3 cells. Both cell types express the EGF receptor and showed decreased junctional conductance in response to EGF. In contrast, EGF did not affect junctional permeability in NIH 3T3 cells, in which EGF receptors were below the detection level (Maldonado et aI., 1988). To examine the possibility that EGF-induced disruption of gjc in T518 cells was due to gross alterations in gap junctional piaques, immunofluorescent microscopy using the connexin43 CT368 peptide antibody was conducted (Figure 3.3; Lau et aI., 1992). Distinct, punctate 62 Cx43 antibody reactions most likely representing gap junctional plaques appeared aggregated and intact for all times examined including the 30 min time point when gjc was greatly reduced. The pattern of immunoreactivity observed in T51 B cells closely resembled gap junctional plaques immunolocalized from frozen sections of liver (Hertzberg and Spray, 1985). Furthermore, phase-contrast microscopy demonstrated that T51 B cells remained intact without any apparent loss of cell-cell contact between adjacent cells following EGF treatment (Figure 3.3; Lau et aI., 1992). These results suggest that EGF's ability to disrupt gjc was unlikely to be due to gross alterations in gap junctional plaques. Receptor autophosphorylation is one consequence of EGF-induced receptor tyrosine kinase activation. In T51B cells, the EGF receptor tyrosine kinase is activated in a rapid and transient manner in response to EGF (Figure 3.2). Maximum levels of activation occurred within 5 min of agonist addition, well before disruption of gjc occurred (Figure 3.1). EGF receptor activation may be the earliest of a cascade of events leading to the disruption of gjc and Cx43 phosphorylation in T51B cells. Maldonado et al. (1988) proposed two possible mechanisms by which EGF could modulate junctional permeability. The first mechanism involves Ca2+. Previous studies have demonstrated growth factor-induced elevation of intracellular Ca2+ concentration in human fibroblasts (Moolenar et aI., 1984). Moreover, Rose and Loewenstein (1975) demonstrated Ca2+-mediated reduction of junctional permeability. While these studies suggest that Ca2+ may mediate EGF-induced disruption of gjc in T51B cells, ionomycin-mediated increases in intracellular Ca2+ levels did not result in reduced gjc (Danesh and Boynton, unpublished observations). Moreover, EGF's ability to disrupt gjc appears to be 63 independent of extracellular Ca2+ levels (Danesh and Boynton, unpublished observations). Taken together, these results suggest that EGF-induced disruption of gjc in T518 cells is unlikely due to the Ca2+ mechanism proposed. The second mechanism proposed by Maldonado et al (1988) suggested that reduced junctional permeability may be due to tyrosine phosphorylation of the connexin protein by receptor tyrosine kinases. Previous studies have correlated reduced junctional permeability in src­ transformed cells with tyrosine phosphorylation of the connexin protein (Crow et aI., 1990, 1992; Swenson et aI., 1990). Whether the tyrosine kinase of pp60src directly phosphorylates connexin or activates a downstream tyrosine kinase has not been determined. EGF treatment of T518 cells resulted in a marked increase in Cx43 phosphorylation (Figure 3.4A; Lau et al., 1992) which occurred in a rapid and transient manner. EGF's ability to stimulate Cx43 phosphorylation inversely correlated with its effects on reducing junctional permeability (compare Figures 3.1 and 3.4A). Phosphoamino acid analysis of Cx43 phosphorylated in response to EGF indicated that the protein was phosphorylated only on serine residues for all time points examined (Figure 3.4-8). In contrast to previous studies with src-transformed cells, these results suggested that the activated EGF receptor itself or another tyrosine kinase(s) is not directly involved but that a serine kinase(s) is acting directly upstream of Cx43. To investigate the possible functional relationship between Cx43 phosphorylation and reduced junctional permeability, dose-dependence of EGF's effects on junctional permeability and Cx43 phosphorylation were examined. Increasing concentrations of EGF resulted in a progressive 64 decrease in gjc and a concomitant increase in Cx43 phosphorylation (Figure 3.5). Therefore, in agreement with previous studies, Cx43 phosphorylation appears to be associated with the disruption of gjc in T51 B cells. The time course of EGF-induced Cx43 phosphorylation demonstrated that following EGF treatment for 3 hr, Cx43 phosphorylation is restored nearly to levels observed in control cells (Figure 3.4A). This occurrence may result from the inactivation of a kinase(s) and/or the activation of a phosphatase(s). A dynamic collaboration between kinases and phosphatases is certainly possible. Okadaic acid, an effective serine/threonine phosphatase inhibitor (Bailojan and Takai, 1988; Haystead et aI., 1989) has previously been utilized as a means to investigate the functional significance of phosphorylation/dephosphorylation events. Okadaic acid exhibits high specificity towards type 2A (pp2A) and type 1 phosphatases (pp1), inhibiting their activities at concentrations of 1 nM and 1 ~M, respectively (Bialojan and Takai, 1988). Okadaic acid at a concentration of 1 nM inhibited the dephosphorylation of Cx43 which typically occurred following 3 hr EGF treatment (Figure 3.5). These results suggested that pp2A may mediate the dephosphorylation of Cx43 observed at the 3 hr time point. Higher concentrations of okadaic acid (1 p-M) resulted in the detachment of cells from the culture dish, therefore, the possible involvement of pp1 on Cx43 dephosphorylation could not be examined. Most notably, okadaic acid's ability to prevent the dephosphorylation of Cx43 was coincident with its ability to inhibit the restoration of gjc observed following 3 hr EGF treatment (Figure 3.5). Taken together, these results demonstrate a strong correlation between the phosphorylation states of Cx43 and modulation of junctional 65 permeability. While these findings are significant, site-directed mutagenesis of the EGF-induced phosphorylation sites will be required to determine the functional significance of EGF's ability to stimulate the serine phosphorylation of Cx43. While the studies in Chapter 3 strongly suggest that EGF-induced disruption of gjc may result from the serine phosphorylation of Cx43, the signalling pathways mediating EGF's effects have not been established. Numerous signalling cascades generated in response to activated growth factor receptor tyrosine kinases have been proposed. Previous studies have characterized TPA- and OAG-induced disruption of gjc (Enomoto and Yamasaki, 1985; Gainer and Murray, 1985; Dotto et al., 1989; Somogyi et aI., 1989; Katoh et aI., 1990). In addition, others have demonstrated TPA-induced phosphorylation of Cx26, Cx43 and MP26, a lens gap junction protein, in vivo (Lampe and Johnson, 1989; Brissette et al., 1991; Oh et aI., 1991; Berthoud et aI., 1992) or direct phosphorylation of Cx32 by purified PKC in vitro (Takeda et aI., 1987; Saez et aI., 1990). Furthermore, EGF has been demonstrated to stimulate the phospholipase C/ phosphatidylinositol pathway ultimately leading to the activation of PKC (Berridge, 1987; Nishizuka, 1988). These earlier studies suggest that EGF may stimulate PKC activity in T51 B cells resulting in the disruption of gjc and serine phosphorylation of Cx43. To examine this possibility, TPA-sensitive PKC was down-modulated with chronic TPA treatment as previously described (Young et al., 1987; Wolfman et al., 1987; Adams and Gullick, 1989) prior to assaying EGF­ induced disruption of gjc and Cx43 phosphorylation. Preincubation of T51 B cells with TPA (100 ng/ml) for 24 hr inhibited the TPA-induced disruption of gjc and Cx43 phosphorylation previously observed (Figures 66 4.1 A and 4.1 B) suggesting that PKC activity has been down-modulated. Direct measurement of PKC activity in T51B cells demonstrated that chronic TPA treatment did indeed down-modulate PKC activity (Figure 4.2). Conversely, chronic TPA treatment did not affect EGF's ability to disrupt gjc and phosphorylate Cx43 (Figure 4.1 A and 4.1B). Phosphoamino acid analysis of Cx43 phosphorylated in response to EGF following chronic TPA treatment demonstrated that only serine residues were phosphorylated for all Cx43 species and time points examined (Figure 4.3). These results indicate that chronic TPA treatment does not induce the activation of a tyrosine or threonine kinase to directly phosphorylate Cx43, but that a serine kinase(s) is involved. Possibly the same serine kinase(s) acting on Cx43 in response to EGF alone may be invovled. Most notably, two-dimensional tryptic phosphopeptide mapping demonstrated that the marked phosphorylation of Cx43 on numerous sites in response to EGF was (with three exceptions) largely unaffected by chronic TPA treatment (compare Figures 4.48 and 4.40). The three missing Cx43 phosphopeptides may possibly contain TPA-sensitive PKC phosphorylation sites which do not appear to be necessary for EGF's effects on Cx43. As mentioned earlier, computer-assisted sequence analysis of Cx43 identified several protein kinase phosphorylation consensus sites including two MAP kinase sites (Alvarez et aI., 1991) as well as sites for PKC, p34Cdc2, casein kinase I and glycogen synthase kinase-3 (Kennelly and Krebs, 1991). Our data suggest that one or more of these other serine kinases may mediate EGF's effects on gjc and Cx43 phosphorylation.

67 The studies in Chapter 4 suggested that EGF is able to modulate gjc and Cx43 phosphorylation by a signalling pathway independent of TPA­ sensitive PKC. As mentioned earlier, the presence of two MAP kinase phosphorylation consensus sites (Alvarez et aI., 1991) as well as EGF's ability to activate MAP kinase independent of PKC activity (Northwood and Davis, 1990) suggested that EGF-induced disruption of gjc and Cx43 phosphorylation may be mediated by MAP kinase. We found that EGF treatment of T51 B cells resulted in the tyrosine phosphorylation and activation of p42 and p44 isoforms of MAP kinase in a time- and dose-dependent manner (Figures 5.1 and 5.2). These results are consistent with previous reports describing MAP kinase as a mitogen­ activated serine/threonine kinase requiring tyrosine and threonine phosphorylation for activity (Ray and Sturgill, 1988b; Rossomando et aI., 1989; Anderson et aI., 1990; Ferrel and Martin, 1990; Boulton et aI., 1991; Payne et aI., 1991). In addition, the kinetics of EGF-induced activation of MAP kinase in T51 B cells was similar to previous studies demonstrating EGF-induced MAP kinase activation in human foreskin fibroblasts, human epithelial carcinoma A431 cells and rat embryonic fibroblastic 3Y1 cells (Gotoh et aI., 1990b; Chao et aI., 1992). EGF-induced stimulation of MAP kinase activity occurred prior to the disruption of gjc and Cx43 phosphorylation which suggested that MAP kinase may mediate those responses. This hypothesis was further substantiated when we demonstrated the ability of purified MAP kinase to phosphorylate Cx43 in vitro (Figure 5.3). Moreover, tryptic phosphopeptide mapping demonstrated that Cx43 was phosphorylated by MAP kinase in vitro on several sites which represented a subset of the sites phosphorylated in

68 vivo in response to EGF (Figure 5.4). These results strongly suggest that Cx43 may be an in vivo substrate of MAP kinase. While previous studies have associated EGF receptor activation to the activation of MAP kinase (Ahn et aI., 1990a,b, 1991), the possible intermediate events involved have only recently become clearer. It has become evident that EGF-stimulated signalling involves, in part, multiple protein interactions and kinase cascades downstream of the activated receptor tyrosine kinase. Upon ligand-induced receptor activation, SH2­ containing Shc proteins associate with the receptor and become tyrosine phosphorylated (Pelicci et aI., 1992). Grb-2, a 23 kDa SH2-containing protein also forms a complex with Shc and the receptor (Rozakis-Adcock et aI., 1992). These protein interactions have been implicated in the activation of Ras in both C. elegans and mammalian HER 14 cells (Clark et aI., 1992; Lowenstein et aI., 1992). Ras activation has been demonstrated to ultimately lead to MAP kinase activation (Wood et aI., 1992; de Vries­ Smits et aI., 1992; Hattori et aI., 1992; Pomerance et aI., 1992) mediated by a complex protein kinase cascade involving Raf-1 ( Dent et aI., 1992; Howe et aI., 1~92; Kyriakis et aI., 1992; Troppmair et aI., 1992; Wood et aI., 1992), possibly a MAP kinase kinase kinase (Gomez et aI., 1992) and MAP kinase kinase (Adams and Parker, 1992; Haystead et aI., 1992; Nakielny et aI., 1992; Matsuda et aI., 1992; Wu et aI., 1992). Thus, this signalling pathway may not only mediate the reduced junctional permeability observed in EGF-treated cells, but also cells transformed by the ras oncogene (Datto et aI., 1989; EI-Fouly et aI., 1989; Brissette et aI., 1991 ). A number of proteins have been identified as substrates of MAP kinase including MBP (Erickson et aI., 1990), MAP-2 (Ray and Sturgill, 69 1987), ribosomal S6 kinase (Sturgill et aI., 1988), c-jun (Alvarez et aI., 1991; Pulverer et aI., 1991), transcription factor p62TCF (Gille et aI., 1992), c-myc (Alvarez et aI., 1991) and the EGF receptor (Countaway et aI., 1989; Northwood et aI., 1991; Takishima et aI., 1991). This study suggests that Cx43 may also be a substrate of MAP kinase. Taken together, the data presented in this study demonstrated that EGF's ability to disrupt gjc in T51B rat liver epithelial cells was correlated with the serine phosphorylation of Cx43. Moreover, the signalling cascade(s) stimulated by EGF does not appear to require TPA-sensitive PKC but appears to be mediated, in part, by MAP kinase. Although this study examines that role of kinases in mediating EGF-induced disruption of gjc and Cx43 phosphorylation, identification and characterization of the possible serine phosphatase(s) involved will also be of interest. In addition, preparation and characterization of phosphorylation site mutants as well as other genetic manipulations will enable us to establish the functional significance of these EGF-induced phosphorylation events. Furthermore, EGF's ability to modulate gjc and Cx43 phosphorylation in T51 B cells provides an excellent system for facilitating further characterization of EGF-stimulated protein kinase signalling pathways involved in regulating critical cellular functions.

70 References

Adams, P.O. and Parker, P.J. (1991) TPA-induced activation of MAP kinase. FEBS Lett. 290:77-82.

Adams, P.O. and Parker, P.J. (1992) Activation of mitogen-activated protein (MAP) kinase by a MAP kinase-kinase. J. BioI. Chern. 267:13135­ 13137.

Adams, J.C. and Gullick, W.J. (1989) Differences in phorbol-ester-induced down-regulation of protein kinase C between cell lines. Biochem. J. 257:905-911.

Ahn, N.G., Weiel, J.E., Chan, C.P. and Krebs, E.G. (1990a) Identification of multiple epidermal growth factor-stimulated protein serine/threonine kinases from Swiss 3T3 cells. J. BioI. Chern. 265:11487-11494.

Ahn, N.G. and Krebs, E.G., (1990b) Evidence for an epidermal growth factor stimulated protein kinase cascade in Swiss 3T3 cells. J. BioI. Chern. 265:11495-11501.

Ahn, N.G., Seger, R., Bratlien, R.L., Diltz, C.D., Tonks, N.K. and Krebs, E.G. (1991) Multiple components in an epidermal growth factor-stimulated protein kinase cascade. J. BioI. Chern. 266:4220-4227.

71 Alvarez, E., Northwood, I.C., Gonzalez, F.A., Latour, D.A., Seth, A., Abate, C., Curran, T. and Davis, R.J. (1991) Pro-Leu-Ser/Thr-Pro is a consensus primary sequence for substrate protein phosphorylation. J. BioI. Chem. 266:15277-15285.

Anderson, N.G., Maller, J.L., Tonks, N.K. and Sturgill, T.W. (1990) Requirement for integration of signals from two distinct phosphorylation pathways for activation of MAP kinase. Nature (London) 343:651-653.

Asamoto, M., Oyamada, M., Aoumari, A.E., Gros, D. and Yamasaki, H. (1991) Molecular mechanisms of TPA-mediated inhibition of gap-junctional intercellular communication: evidence for action on the assembly or function but not the expression of connexin 43 in rat liver epithelial cells. Mol. Carcinogenesis 4:322-327.

Atkinson, M.M., Menko, A.S., Johnson, R.G., Sheppard, J.R. and Sheridan, J.D. (1981) Rapid and reversible reduction of junctional permeability in cells infected with a temperature-sensitive mutant of avian sarcoma virus. J. Cell BioI. 91 :573-578.

Atkinson, M.M. and Sheridan, J.D. (1988) Altered junctional permeability between cells transformed by v-res, v-mos, or v-src. Am. J. Physiol. 255:674-683.

72 Azarnia, R. and Loewenstein, W.R. (1984) Intercellular communication and the control of growth. X. Alteration of junctional permeability by the src gene. A study with temperature-sensitive mutant Rous sarcoma virus. J. Membr. BioI. 82:191-205.

Azarnia, R. and Loewenstein, W.R. (1987) Polyomavirus middle T antigen downregulates junctional cell-to-cell communication. Mol. Cell. BioI. 7:946­ 950.

Azarnia, R., Reddy, S., Kmiecik, R.E., Shalloway, D. and Loewenstein, W.R. (1988) The cellular src gene product regulates junctional cell-to-cell communication. Science 239:398-401.

Bell, R.M. (1986) Protein kinase C activation by diacylglycerol second messengers. Cell 45:631-632.

Berridge, M.J. (1987) Inositol triphosphate and diacylglycerol: Two interacting second messengers. Annu. Rev. Biochem. 56:156-194.

Berthoud, V,M., Ledbetter, M.S., Hertzberg, E.L. and Saez, J.C. (1992) Connexin43 in MDCK cells: regulation by a tumor-promoting phorbol ester and Ca2+. Eur. J. Cell BioI. 57:40-50.

Beyer, E.C., Paul, D.L. and Goodenough, D.A. (1990) Connexin family of gap junction proteins. J. Membr. BioI. 116:187-194.

73 Bialojan, C. and Takai, A. (1988) Inhibitory effect of a marine-sponge toxin, okadaic acid, on protein phosphatases. Biochem. J. 256:283-290.

Bignami, M., Rosa, S., Falcone, G., Tato, F., Katoh, F. and Yamasaki, H. (1988) Specific viral oncogenes cause differential effects on cell to cell communication, relevant to the suppression of the transformed phenotype by normal cells. Mol. Carcinogenesis 1:67-75.

Blumberg, P.M. (1988) Protein kinase C as the receptor for the phorbol ester tumor promoters: Sixth Rhoads Memorial Award lecture. Cancer Res. 48:1-8.

Bolen, J.B., Thiele, C.J., Israel, M.A., Yonemoto, W., Lipsich, L.A. and Brugge, J.S. (1984) Enhancement of cellular src gene product associated tyrosyl kinase activity following polyoma virus infection and transformation. Cell 38:767-777.

Boulton, T.G., Gregory, J.S. and Cobb, M.H. (1991 a) Purification and properties of extracellular signal-regulated kinase 1, an insulin-stimulated microtubule associated protein 1 kinase. Biochemistry 30:278-286.

Boulton, T.G., Nye, S.H., Robbins, D.J., Ip, N.Y., Radziejewska, E., Morgenbesser, S.D., DePinho, R.A., Panayotatos, N., Cobb, M.H. and Yancopoulos, G.D. (1991 b) ERKs: A family of protein-serine/threonine kinases that are activated and tyrosine phosphorylated in response to insulin and NGF. Cell 65:663-675.

74 Boyle, W.J., van der Geer, P. and Hunter, T. (1991) Phosphopeptide mapping and phosphoamino acid analysis by two-dimensional separation on thin-layer cellulose plates. Methods Enzymo!. 201: 110-149.

Brissette, J.L., Kumar, N.M., Gilula, N.B. and Dotto, G.P. (1991) The tumor promoter 12-0-tetradecanoylphorbol-13-acetate and the ras oncogene modulate expression and phosphorylation of gap junction proteins. Mol. Cell. BioI. 11 : 5364-5371.

Burns, D.J. and Bell, R.M. (1991) Protein kinase C contains two phorbol ester binding domains. J. BioI. Chem. 266: 18330-18338.

Carpenter, G. and Cohen, S. (1990) Epidermal growth factor. J. BioI. Chem. 265:7709-7712.

Chang, C-C., Trosko, J.E., Kung, H-J., Bombrik, D. and Matsumura, R. (1985) Potential role of the src gene product in inhibition of gap junctional communication in NIH3T3 cells. Proc. Natl. Acad. Sci. USA 82:5360-5364.

Chao, T-S, 0., Byron, K.L., Lee, K-M, Villereal, M. and Rosner, M.R. (1992) Activation of MAP kinases by calcium-dependent and calcium­ independent pathways. J. BioI. Chem. 267:19876-19883.

Clark, S.G., Stern, M.J. and Horvitz, H.R. (1992) C. elegans cell-signaling gene sem-5 encodes a protein with SH2 and SH3 domains. Nature (London) 356:340-344.

7S Cochet, C., Filhol, 0., Payrastre, B., Hunter, T. and Gill, G.N. (1991) Interaction between the epidermal growth factor receptor and phosphoinositide kinase. J. BioI. Chern. 266:637-644.

Countaway, J.L., Northwood, I.C. and Davis, R.J. (1988) Mechanism of phosphorylation of the epidermal growth factor receptor at threonine 669. J. BioI. Chern. 264:10828-10835.

Courchesne, W.E., Kunisawa, R. and Thorner, J. (1989) aA putative protein kinase overcomes phermone-induced arrest of cell cycling in S. cerevisiae. Cell 58:1107-1119.

Crow, D.S., Beyer, E.C., Paul, D.C., Kobe, 8.S. and Lau, A.F. (1990) Phosphorylation of connexin43 gap junction protein in uninfected and Rous sarcoma virus-transformed mammalian fibroblasts. Mol. Cell. BioI. 10:1754-1763.

Crow, D.S., Kurata, W.E. and Lau, A.F. (1992) Phosphorylation of connexin43 in cells containing mutant src oncogenes. Oncogene 7:999­ 1003. de Vries-Smits, A.M.M., Bergering, B.M.Th., Leevers, S.J., Marshall, C.J. and Bos, J.L. (1992) Involvement of p21 ras in activation of extracellular signal-regulated kinase 2. Nature (London) 357:602-604.

76 Dent, P., Haser, W., Haystead, T.A.J., Vincent, L., Roberts, T.M. and Sturgill, T.W. (1992) Activation of mitogen-activated protein kinase kinase by v-Raf in NIH 3T3 cells and in vitro. Science 257:1404-1407.

Dermietzel, R., Hwang, T.K. and Spray, D.S. (1990) The gap junction family: Structure, function and chemistry. Anal. Embryol. 182:517-528.

Dotto, G.P., EI-Fouly, M.H., Nelson, C. and Trosko, J.E. (1989~ Similar and synergistic inhibition of gap-junctional communication by {as transformation and tumor promoter treatments of mouse primary keratinocytes. Oncogene 4:637-641.

EI-Fouly, M., Trosko, J.E., Chang, C.C. and Warren, S.R. (1989) Potential role of the human Ha-ras oncogene in the inhibition of gap junctional intercellular communication. Mol. Carcinogenesis 2:131-135.

Elion, E.A., Grisafi, P.L. and Fink, G.R. (1990) FUS3 encodes a cdc2/CDC28-related kinase required for the transition from mitosis into conjugation Cell 60:649-664.

Enomoto, T. and Yamasaki, H. (1985) Rapid inhibition of intercellular communication between BALB/c 3T3 cells by diacylglycerol, a possible endogenous functional analogue of phorbol esters. Cancer Res. 45:3706­ 3710.

77 Erickson, A.K., Payne, D.M., Martino, P.A., Rossomando, A.J., Shabanowitz, J., Weber, M.J., Hunt, D.F. and Sturgill, T.W. (1990) Identification by mass spectrometry of threonine 97 in bovine myelin basic protein as a specific phosphorylation site for mitogen-activated protein kinase. J. BioI. Chem. 265:19728-19735.

Ferrell, J.E. and Martin, G.S. (1990) Identification of a 42-kilodalton phosphotyrosyl protein as a serine (threonine) protein kinase by renaturation. Mol. Cell. BioI. 10:3020-3026.

Flagg-Newton, J.L., Dahl, G. and Loewenstein, W.R. (1981) and cyclic AMP:1. Up regulation of junctional membrane permeability and junction membrane particles by administration of cyclic or phosphodiesterase inhibitor. J. Membr. BioI. 63:105-107

Furshpan, E.J. and Potter, D.O. (1959) Transmission at the giant motor synapses of the crayfish. J. Physiol. 145:289-325.

Gainer, H.St. C. and Murray, A.W. (1985) Diacylglycerol inhibits gap junctional communication in cultured epidermal cells: Evidence for a role of protein kinase C. Biochem. Biophys. Res. Commun. 126:1109-1113.

Ganong, B.A., loomis, C.A., Hannun, Y.A. and Bell, R.M. (1986) Specificity and mechanism of protein kinase C activation by sn-1,2-diacylglycerols. Proc. Natl. Acad. Sci. USA 83:1184-1188.

78 Garfield, R.E., Sims, S., Hannan, M.S. and Daniel, E.E. (1977) Gap junctions: their presence and necessity in myometrium during parturition. Science 198:958-961.

Gille, H., Sharrocks, A.D. and Shaw, P.E. (1992) Phosphorylation of transcription factor p62TCF by MAP kinase stimulates ternary complex formation at c-tos promoter. Nature (London) 358:414-417.

Glenney Jr., J.R., (1992) Tyrosine-phosphorylated proteins: Mediators of from the tyrosine kinases. Biochimica et Biophysica Acta. 1143:113-127.

Gomez, N., Traverse, S. and Cohen, P. (1992) Identification of a MAP kinase kinase kinase in phaeochromocytoma (PC12) cells. FEBS Lett. 314:417-421.

Gonzalez, F.A., Raden, D.L., Rigby, M.R. and Davis, R.J. (1992) Heterogeneous expression of four MAP kinase isoforms in human tissues. FEBS Lett. 304:170-178.

Gotoh, Y., Nishida, E., Yamashita, T., Hoshi, M., Kawakami, M. and Sakai, H. (1990a) Microtubule-associated protein (MAP) kinase activated by nerve growth factor and epidermal growth factor in PC12 cells. Eur. J. Biochem. 193:661-669.

79 Gotoh, Y., Nishida, E. and Sakai, H. (1990b) Okadaic acid activates microtubule-associated protein kinase in quiescent fibroblastic cells. Eur. J. Biochem. 193:671-674.

Haefliger, J-A., Bruzzone, R., Jenkins, N.A., Gilbert, D.J., Copeland, N.G. and Paul, D.L. (1992) Four novel members of the connexin family of gap junction proteins. J. BioI. Chem. 267:2057-2064.

Hattori, S., Fukuda, M., Yamashita, T., Nakamura, S., Gotoh, Y. and Nishida, E. (1992) Activation of mitogen-activated protein kinase and its activator by ras in intact cells and in a cell-free system. J. BioI. Chern. 267:20346-20351.

Haystead, T.A.J., Sim, A.T.R., Carling, D., Honner, R.C., Tsukitani, Y., Cohen, P. and Hardie, D.G. (1989) Effects of the tumour promoter okadaic acid on intracellular protein phosphorylation and metabolism. Nature 337:78-81.

Haystead, T.A.J., Dent, P., Wu, J., Haystead, C.M.M. and Sturgill, T.W. (1992) Ordered phosphorylation of p42mapk by MAP kinase kinase. FEBS Lett. 306:17-22.

Helin, K., Velu, T., Martin, P., Vass, w.e., Allevato, G., Lowy, D.R. and Beguinot, L. (1991) The biological activity of the human epidermal growth factor receptor is positively regulated by its C-terminal tyrosines. Oncogene 6:825-832.

80 Hernandez-Sotomayor, S.M.T. and Carpenter, G. (1992) Epidermal growth factor receptor: Elements of intracellular communication. J. Membr. BioI. 128:81-89.

Hertzberg, E.L. and Spray, D.C. (1985) Studies of gap junctions: Biochemical analysis and use of antibody probes. In: Gap Junctions (Bennett, M.V.L. and Spray, D.C., eds.) Cold Spring Harbor Laboratory, New York 57-65.

Hill, T.D., Campos-Gonzalez, R., Kindmark, H. and Boynton, A.L. (1988) Inhibition of inositol trisphosphate--stimulated calcium mobilization by calmodulin antagonists in rat liver epithelial cells. J. BioI. Chem. 263:16479-16484.

Howe, L.R., Lewers, S.J., Gomez, N., Nakielny, S., Cohen, P. and Marshall, C.J. (1992) Activation of the MAP kinase pathway by the protein kinase rat. Cell 71 :335-342.

Hsu, C-Y, J., Hurwitz, D.R., Mervie, M. and Zilberstein, A. (1991) Autophosphorylation of the intracellular domain of the epidermal growth factor receptor results in different effects on its tyrosine kinase activity with various peptide substrates. J. BioI. Chem. 266:603-608.

Kameshita, I. and Fujisawa, H. (1989) A sensitive method for detection of calmodulin-dependent protein kinase II activity in sodium dodecyl sulfate­ polyacrylamide gel. Anal. Biochem. 183:139-143.

81 Kamps, M.P. and Sefton, 8.M. (1989) Acid and base hydrolysis of phosphoproteins bound to Immobilon facilitates analysis of phosphoamino acids in gel-fractionated proteins. Anal. Biochem. 176:22-27.

Katoh, F., Fitzgerald, D.J., Giroldi, L., Fujiki, H., Sugimura, T. and Yamasaki, H. (1990) Okadaic acid and phorbol esters: Comparative effects of these tumor promoters on cell transformation, intercellular communication and differentiation in vitro. Jpn. J. Cancer Res. 81 :590­ 597.

Kazlauskas, A. and Cooper, J.A. (1988) Protein kinase C mediates platelet-derived growth factor-induced tyrosine phoshorylation of p42. J. Cell BioI. 106:1395-1402.

Kennelly, P.J. and Krebs, E.G. (1991) Consensus sequences as substrate specificity determinants for protein kinases and protein phosphatases. J. BioI. Chem. 266:15555-15558.

Klaunig, J.E. and Ruch, R.J. (1990) Biology of disease. Role of inhibition of intercellular communication in carcinogenesis. Lab. Invest. 62:135-146.

Koch, C.A., Anderson, D., Moran, M.F., Ellis, C. and Pawson, T. (1991) SH2 and SH3 domains: elements that control interactions of cytoplasmic signalling proteins. Science 252:668-674.

82 Konishi, N., Donovan, P.J. and Ward, J.M. (1990) Differential effects of renal carcinogens and tumor promoters on growth promotion and inhibition of gap junctional communication in two rat renal epithelial cell lines. Carcinogenesis 11 :903-908.

Kurata, W.E. and Lau, A.F. (1993) Manuscript in preparation.

Kyriakis, J.M., Brautigan, D.L., Ingebritsen, T.S., and Avruch, J. (1991) pp54 microtubule-associated protein 2 kinase requires both tyrosine and serine/threonine phosphorylation for activity. J. BioI. Chern. 266:10043­ 10046.

Kyriakis, J.M., App, H., Zhang, X-f, Banerjee, P., Brautigan, D.L., Rapp, U.R. and Avruch, J. (1992) Raf-1 activates MAP kinase-kinase. Nature (London) 358:417-421.

Laemmli, U.K. (1970) Cleavage of stuctural proteins during the assembly of the head bacteriophage T4. Nature 227:680-685.

Laird, D.W., Puranam, K.L. and Revel, J.P. (1991) Turnover and phosphorylation dynamics of connexin43 gap junction protein in cultured cardiac myocytes. Biochem. J. 273:67-72.

L'Allemain, G., Sturgill, T.W. and Weber, M.J. (1991) Defective regulation of mitogen-activated protein kinase activity in a 3T3 cell variant mitogenically nonresponsive to tetradecanoylphorbol acetate. Mol. Cell. BioI. 11:1002-1008. 83 L'Allemain, G., Her, J-H, Wu, J., Sturgill, T.W. and Weber, M.J. (1992) Growth factor-induced activation of a kinase activity which causes regulatory phosphorylation of p42/microtubule-associated protein kinase. Mol. Cell. BioI. 12:2222-2229.

Lampe, P.O. and Johnson, R.G. (1989) Phosphorylation of MP26, a lens junction protein, is enhanced by activators of protein kinase C. J. Membr. BioI. 107:145-155.

Lau, A.F., Hatch-Pigott, V. and Crow, D.S. (1991) Evidence that heart connexin43 is a phosphoprotein. J. Mol. Cell Cardiol. 23:659-663.

Lau, A.F., Kanemitsu, M.Y., Kurata, W.E., Danesh, S. and Boynton, A.L. (1992) Epidermal growth factor disrupts gap-junctional communication and induces phosphorylation of connexin43 on serine. Mol. BioI. Cell 3:865­ 874.

Loewenstein, W.R. (1981) Junctional intercellular communication: The cell­ to-cell . Physio!. Rev. 61 :829-913.

Loewenstein, W.R. and Rose, B. (1992) The cell-cell channel in the control of growth. Seminars in cell BioI. 3:59-79.

Loo, L., Kurata, W.E., Berestecky, J. and Lau, A.F. (1993) Manuscript in preparation.

84 Lowenstein, E.J., Daly, R.J., Batzer, A.G., Li, W., Margoli, B., Lammers, R, Ullrich, A., Skolnik, E.Y., Bar-Sagi, D. and Schlessinger, J. (1992) The SH2 and SH3 domain-containing protein GRB2 links receptor tyrosine kinases to ras signaling. Cell 70:431-442.

Madhukar, B.V., Oh, S.Y., Chang, C.C., Wade, M. and Trosko, J.E. (1989) Altered regulation of intercellular communication by epidermal growth factor, transforming growth factor - beta and peptide hormones in normal keratinocytes. Carcinogenesis 10:13-20.

Maldonado, P.E., Rose, B. and Loewenstein, W.R. (1988) Growth factors modulate junctional cell-to-cell communication. J. Membr. BioI. 106:203­ 210.

Margolis, B., Li, N., Koch, A., Mohammadi, M., Hurzitz, D.R., Zilberstein, A., Ullrich, A., Pawson, R. and Schlessinger, J. (1990) The tyrosine phosphorylated carboxyterminus of the EGF receptor is a binding site for GAP and PLC-gamma. EMBO J. 9:4375-4380.

Margolis, B. (1992) Proteins with SH2 domains: Transducers in the tyrosine kinase signalling pathway. Cell Growth Differentiation 3:73-80.

Matsuda, S., Kosako, H., Takenaka, K., Moriyama, K., Sakai, H., Akiyama, T., Gotoh, Y. and Nishida, E. (1992) Xenopus MAP kinase activator: identification and function as a key intermediate in the phosphorylation cascade. EMBO J. 11:973-982.

85 Matsudaira, P. (1987) Sequence from picomole quantities of proteins electroblotted onto polyvinyiidene difluoride membranes. J. BioI. Chern. 262:10035-10038.

Mehta, P.P., Bertram, J.S. and Loewenstein, W.R. (1986) Growth inhibition of transformed cells correlates with their junctional communication with normal cells. Cell 44:187-196.

Mehta, P.P., Hotz-Wagenblatt, A., Rose, B., Shalloway, D. and Loewenstein, W.R. (1991) Incorporation of the gene for a cell-cell channel protein into transformed cells leads to normalization of growth. J. Membrane BioI. 124:207-225.

Meier, K.E., Weiel, J.E., Bloom, T.J. and Krebs, E.G. (1990) Regulation of S6 kinase activity in Madin-Darby canine kidney renal epithelial cells. J. BioI. Chern. 265:4635-4645.

Meier, K.E., Licciardi, K.A., Haystead, T.A.J. and Krebs, E.G. (1991) Activation of messenger-independent protein kinases in wild-type and phorbol ester-resistant EL4 Thymoma cells. J. BioI. Chern. 266:1914­ 1920.

Moolenar, W., Tertoolen, L. and de Laat, S. (1984) Growth factors immediately raise cytoplasmic free Ca2+ in human fibroblasts. J. BioI. Chern. 259:8066-8069.

86 Moran, M.F., Koch, C.A., Anderson, D., Ellis, C., England, L., Martin, G.S. and Pawson, T. (1990) Src homology region 2 domains direct protein­ protein interactions in signal transduction. Proc. Natl. Acad. Sci. USA 87:8622-8626.

Moreno, A.P., Fishman, G.!. and Spray, D.C. (1992) Phosphorylation shifts unitary conductance and modifies voltage dependent kinetics of human connexin43 gap junction channels. Biophys. J. 62:51-53.

Musil, L.S., Cunningham, B.A., Edelman, G.M. and Goodenough, D.A. (1990) Differential phosphorylation of the gap junction protein connexin43 in junctional communication-competent and -deficient cell lines. J. Cell BioI. 111:2077-2088.

Nakamura, H., Kishi, Y., Pajares, M.A. and Rando, A.R. (1989) Structural basis of protein kinase C activation by tumor promoters. Proc. Natl. Acad. Sci. USA 86:9672-9676.

Nakielny, S., Cohen, P., Wu, J. and Sturgill, T.W. (1992) MAP kinase activator from insulin-stimulated skeletal muscle is a protein threonine/tyrosine kinase. EMBO J. 11 :2123-2129.

Nel, A.E., Hanekom, C., Rheeder, A., Williams, K., Pollack, S., Katz, R. and Landreth, G.E. (1990) Stimulation of MAP-2 kinase activity in T lymphocytes by anti-CD3 or anti-Ti monoclonal antibody is partially dependent on protein kinase C. J. Immuno!. 144:2683-2689.

87 Niedel, J.E. and Blackshear, P.J. (1986) Protein kinase C In: Receptor biochemistry and methodology. Phosphoinositides and receptor mechanisms (Putney Jr., J.W., ed.) Liss, New York 7:47-88. Nishizuka, Y. (1986) Studies and perspectives of protein kinase C. Science 233:305-312.

Nishizuka, Y. (1988) The molecular heterogeneity of protein kinase C and its implications for cellular regulation. Nature 334:661-665.

Northwood, I.C. and Davis, R.J. (1990) Signal transduction by the epidermal growth factor receptor after functional desensitization of the receptor tyrosine protein kinase activity. Proc. Natl. Acad. Sci. USA 87:6107-6111.

Northwood, I.C., Gonzalez, F.A., Wartmann, M., Raden, D.L. and Davis, R.J. (1991) Isolation and characterization of two growth factor-stimulated protein kinases that phosphorylate the epidermal growth factor receptor at threonine 669. J. BioI. Chern. 266:15266-15276.

Oh, S.Y., Grupen, C.G. and Murray, A.W. (1991) Phorbol ester induces phosphorylation and down-regulation of connexin 43 in WB cells. Biochim. Biophys. Acta 1094:243-245.

Parker, P.J., Coussens, L., Totty, N., Rhee, L., Young, S., Chen, E., Stabel, S., Waterfield, M.D. and Ullrich, A. (1986) The complete primary structure of protein kinase C - the major phorbol ester receptor. Science 233:853-866.

88 Paulson, A.F. and Atkinson, M.M. (1989) Interplay of ras and TPA in reducing cell-cell communication. J. Cell BioI. 109:48a. Pawson, T and Gish, G.D. (1992) SH2 and SH3 domains: From structure to function. Cell 71 :359-362.

Payne, D.M., Rossomando, A.J., Martino, P. Erickson, A.K., Her, J-H, Shabanowitz, J., Hunt, D.F., Weber, M.J. and Sturgill, T.W. (1991) Identification of the regulatory phosphorylation sites in pp42/mitogen­ activated protein kinase (MAP kinase). EMBO J. 10:885-892.

Pelech, S.L. and Sanghera, J.S. (1992) Mitogen-activated protein kinases: versatile transducers for . Trends in Biochem. Sci. 17:233­ 238.

Pelicci, G., Lanfrancone, L., Grignani, F., McGlade, J., Cavallo, F., Forni, G., Nicoletti, F., Grignani, F., Pawson, T. and Pelicci, P.G. (1992) A novel transforming protein (SHC) with an SH2 domain is implicated in mitogenic signal transduction. Cell 70:93-104.

Peracchia, C. and Girsch, S.J. (1985) Functional modulation of cell coupling: evidence for a calmodulin-driven channel gate. Am. J. Physio!. 248:H765-H782.

Peracchia, C. (1988) The calmodulin hypothesis for gap junction regulation six years later. In: Gap junctions (Hertzberg, E.L. and Johnson, R.G., eds) Alan R. Uss, Inc. New York 7:267-282.

89 Pomerance, M., Schweighoffer, F., Tocque, B. and Pierre, M. (1992) Stimulation of mitogen-activated protein kinase by oncogenic ras p21 in Xenopus oocytes. J. BioI. Chem. 267:16155-16160.

Pulverer, B.J., Kyriakis, J.M. Avruch, J., Nikolakaki, E. and Woodgett, J.R. (1991) Phosphorylation of c-jun mediated by MAP kinases. Nature (London) 353:670-674.

Ramon, F. and Rivera, A. (1986) Gap junction channel modulation: A physiological viewpoint. 48:127-153.

Ray, L.B. and Sturgill, T.W. (1987) Rapid stimulation by insulin of a serine/threonine kinase in 3T3-L1 adipocytes that phosphorylates microtubule-associated protein 2 in vitro. Proc. Natl. Acad. Sci. USA 84: 1502-1506.

Ray, L.B. and Sturgill, T.W. (1988a) Characterization of insulin-stimulated microtubule-associated protein kinase. J. BioI. Chem. 263:12721-12727.

Ray, L.B. and Sturgill, T.W. (1988b) Insulin-stimulated microtubule­ associated protein kinase is phosphorylated on tyrosine and threonine in vivo. Proc. Nat!. Acad. Sci. USA 85: 3753-3757.

Risek, B., Guthrie, S., Kumar, N. and Gilula, N.B. (1990) Modulation of gap junction transcription and protein expression during pregnancy in the rat. J. Cell BioI. 110:269-282.

90 Robertson, J.D., Bodenheimer, T.S. and Stage, D.E. (1963) The ultrastructure of Mauthner cell synapses and nodes in goldfish brains. J. Cell. BioI. 19:159-199.

Roesler, W.J., Vandenbark, G.R. and Hansen, R.W. (1988) Cyclic AMP and the induction of eukaryotic gene transcription. J. BioI. Chern. 263:9063-9066.

Rose, B. and Loewenstein, W.R. (1975) Permeability of cell junction depends on local cytoplasmic calcium activity. Nature (London) 254:250­ 252.

Rossomando, A.J., Payne, D.M., Weber, M.J. and Sturgill, T.W. (1989) Evidence that pp42, a major tyrosine kinase target protein, is a mitogen­ activated serine/threonine protein kinase. Proc. Natl. Acad. Sci. USA 86:6940-6943.

Rozakis-Adcock, M., McGlade, J., Mbamalu, G., Pelicci, G., Daly, R., Li, W., Batzer, A., Thomas, S., Brugge, J., Pelicci, P.G., Schlessinger, J. and Pawson, T. (1992) Association of the Shc and Grb/Sem5 SH2-containing proteins is implicated in activation of the Ras pathway by tyrosine kinases. Nature (London) 360:689-692.

Saez, J.e., Spray, D.C., Nairn, A.C., Hertzberg, E.L., Greengard, P. and Bennett, M.V.L. (1986) cAMP increases junctional conductance and stimulates phosphorylation of the 27 kDa principal gap junction polypeptide. Proc. Natl. Acac. Sci. USA 83:2473-2477. 91 Saez, J.C., Nairn, A.C., Czernik, A.J., Spray, D.C., Hertzberg, E.L. and Greengard, P. (1990) Phosphorylation of connexin32, a hepatocyte gap­ junction protein by cAMP-dependent protein kinase, protein kinase C and Ca2+/calmodulin-dependent protein kinase II. Eur. J. Biochem. 192:263­ 273.

Sanghera, J.S., Paddon, H.B. and Pelech, S.L. (1991) Role of protein phosphorylation in the maturation-induced activation of a myelin basic protein kinase from sea star oocytes. J. BioI. Chern. 266:6700-6707.

Sharkey, N.A., Leach, K.L. and Blumberg, P.M. (1984) Competitive inhibition by diacylglycerol of specific phorbol ester binding. Proc. Nat!. Acad. Sci. USA 81 :607-610.

Somogyi, R., Batzer, A. and Kolb, H.A. (1989) Inhibition of electrical coupling in pairs of murine pancreatic acinar cells by OAG and isolated protein kinase C. J. Membr. BioI. 108:273-282.

Sorkin, A., Water, C., Overholfer, K.A. and Carpenter, G. (1991) Multiple autophosphorylation site mutations of the epidermal growth factor receptor. J. BioI. Chern. 266:8355-8352.

Spray, D.C., Harris, A.L. and Bennett, M.V.L. (1981) Gap junctional conductance is a simple and sensitive function of intracellular pH. Science 211 :712-715.

92 Spray, D.C., White, R.L., Verselis, V. and Bennett, M.V.L. (1985) General and comparative physiology of gap junction channels. In: Gap junctions (Bennett, M.V.L. and Spray, D.C. eds.) Cold Spring Harbor Laboratory, New York 139-153.

Spray, D.C., Saez, J.e. and Hertzberg, E.L. (1988) Gap junctions between hepatocytes: Structural and regulatory features. In: The Liver:. Biology and pathology ed.2 (Arias, I.M., Jakoby, W.B., Popper, H., Schachler, D. and Shafritz, D.A., eds) Raven Press, New York 851-865.

Stagg, A.B. and Fletcher, W.H. (1990) The hormone-induced regulation of contact-dependent cell-cell communication by phosphorylation. Endo. Rev. 11 :302-325.

Sturgill, T.W., Ray, L.B., Erickson, E. and Maller, J. (1988) Insulin­ stimulated MAP-2 kinase phosphorylates and activates ribosomal protein S6 kinase II. Nature (London) 334:715-718.

Swenson, K.I., Piwnica-Worms, H., McNamee, H. and Paul, D.L. (1990) Tyrosine phosphorylation of the gap junction protein connexin43 is required for the pp60v-src-induced inhibition of communication. Cell Regul. 1:989-1002.

Takeda, A., Hashimoto, E., Yamamura, H. and Shimazu, T. (1987) Phosphorylation of liver gap junction protein by protein kinase C. FEBS Lett. 210:169-172.

93 Takishima, K., Griswold-Prenner, I., Ingebritsen, T. and Rosner, M.R. (1991) Epidermal growth factor (EGF) receptor T669 peptide kinase from 3T3-L1 cells is an EGF-stimulated "MAP" kinase. Proc. Natl. Acad. Sci. USA 88:2520-2524.

Thomas, S.M., DeMarco, M., D'Arcangelo, G., Halegona, S. and Brugge, J.S. (1992) Ras is essential for nerve growth factor- and phorbol ester­ induced tyrosine phosphorylation of MAP kinases. Cell 68:1031-1040.

Tobe, K., Kadowaki, T., Tamemoto, H., Ueki, K., Hara, K., Koshi, 0., Momomura, K., Gotoh, Y., Nishida, E., Akanuma, Y., Yazaki, Y. and Kasuga, M. (1991) Insulin and 12-0-tetradecanoylphorbol-13-acetate activation of two immunologically distinct myelin basic protein/microtubule­ associated protein 2 (MBP/MAP2) kinases via de Novo phosphorylation of threonine and tyrosine residues. J. BioI. Chem. 266:24793-24803.

Troppmair, J., Bruder, J.R., App, H., Cai, H., Liptak, L., Szeberenyi, J., Cooper, G.M. and Rapp, U.R. (1992) Ras controls coupling of growth factor receptors and protein kinase C in the membrane to Raf-1 and B-Raf protein serine kinases in the cytosol. Oncogene 7:1867-1873.

Vila, J. and Weber, M.J. (1988) Mitogen-stimulated tyrosine phosphorylation of a 42-kD cellular protein: Evidence for a protein kinase C requirement. J. Cell. Physiol. 135:285-292.

94 Wang, H-C.R. and Erikson, R.L. (1992) Activation and protein serine/threonine kinases p42, p63 and p87 in Rous sarcoma virus­ transformed cells. Mol. BioI. Cell 3:1329-1337.

Warner, A. (1988) The gap junction. J. Cell Sci. 89:1-7.

Weiel, J.E., Ahn, N.G., Seger, R. and Krebs, E.G. (1990) Communication between protein tyrosine and protein serine/threonine phosphorylation. In: The Biology and Medicine of Signal Transduction. (Nishizuka et a!. eds.) Raven Press, New York 182-195.

Wolfman, A., Wingrove, T.G., Blackshear, P.J. and Macara, I.G. (1987) Down-regulation of protein kinase C and of an endogenous 80-kDa substrate in transformed fibroblasts. J. BioI. Chem. 262: 16546-16552.

Wood, K.W., Sarnecki, C., Roberts, T.M. and Blenis, J. (1992) ras mediates nerve growth factor receptor modulation of three signal­ transducing protein kinases: MAP kinase, Raf-1 and RSK. Cell 68:1041­ 1050.

Wu, J., Michel, H., Rossomando, A., Haystead, T., Shabanowitz, J., Hunt, D.F. and Sturgill, T.W. (1992) Renaturation and partial peptide sequencing of mitogen-activated protein kinase (MAP kinase) activator from rabbit skeletal muscle. Biochem. J. 285:701-705.

95 Yamasaki, H. (1990) Gap junctional intercellular communication and carcinogenesis. Carcinogenesis 11:1051-1058. Yarden, Y. and Ullrich, A. (1988) Growth factor receptor tyrosine kinases. Annu. Rev. Biochem. 57:443-478.

Young, S., Parker, P.J., Ullrich, A. and Stabel, S. (1987) Down-regulation of protein kinase C is due to an increased rate of degradation. Biochem. J. 244:775-779.

96