Experimental Cell Research 336 (2015) 223–231

Contents lists available at ScienceDirect

Experimental Cell Research journal homepage: www.elsevier.com/locate/yexcr

Differential effects of polyoma virus middle tumor antigen mutants upon gap junctional, intercellular communication Mulu Geletu 1, Stephanie Guy 3, Samantha Greer 2,3, Leda Raptis n Department of Biomedical and Molecular Sciences and Department of Pathology and Molecular Medicine, Queen's University, Kingston, Ontario, Canada K7L 3N6

art ic l e i nf o

a b s t r a c t

Article history: Received 29 January 2015 Received in revised form 7 July 2015 Accepted 13 July 2015 Available online 14 July 2015

Gap junctions are channels that connect the cytoplasm of adjacent cells. Oncogenes such as the middle Tumor antigen of polyoma virus (mT) are known to suppress gap junctional, intercellular communication (GJIC). mT associates with and is tyrosine-phosphorylated by cSrc family members. Specific mT phosphotyrosines provide docking sites for the phosphotyrosine binding domain of Shc (mT-tyr250) or the SH2 domain of the regulatory subunit of the phosphatidylinositol-3 kinase (PI3k, mT-tyr315). Binding results in the activation of their downstream signaling cascades, Ras/Raf/Erk and PI3 kinase/Akt, respectively, both of which are needed for full neoplastic transformation. To examine the effect of mT-initiated pathways upon gap junctional communication, GJIC was quantitated in rat liver epithelial T51B cells expressing mT-mutants, using a novel technique of in situ electroporation. The results demonstrate for the first time that, although even low levels of wild-type mT are sufficient to interrupt gap junctional communication, GJIC suppression still requires an intact tyr-250 site, that is activation of the Ras pathway. In sharp contrast, activation of the PI3k pathway is not required for GJIC suppression, indicating that GJIC suppression is independent of full neoplastic conversion and the concomitant morphological changes. Interestingly, expression of a constitutively active, myristylated form of the catalytic subunit of PI3k, p110, or the constitutively active mutants E545K and H1047R increased GJIC, while pharmacological inhibition of PI3k eliminated communication. Therefore, although PI3k is growth promoting and in an activated form it can act as an oncogene, it actually plays a positive role upon gap junctional, intercellular communication. & 2015 Elsevier Inc. All rights reserved.

Keywords: Polyoma virus transformation Gap junctions Src Phosphatidylinositol-3 kinase Ras pathway Electroporation in situ

1. Introduction Gap junctions are aqueous channels connecting the cytoplasm of adjacent cells, that permit the direct transfer of small molecules and ions [41]. These channels are composed of integral membrane proteins, the connexin family. Gap junctions are created by the

Abbreviations: Erk1/2, extracellular signal regulated kinase 1 and 2; GJIC, gap junctional, intercellular communication; LY, Lucifer yellow; mT, middle tumor antigen of polyoma virus; myr-p110, myristylated, p110 catalytic subunit of PI3k; PI3k, phosphatidylinositol-3 kinase; p110, catalytic subunit of PI3k; PKC, protein kinase C; PKA, cAMP-dependent kinase; PLCγ, phospholipase-C gamma; PTB, phosphotyrosine binding domain; Ras, rat sarcoma; SH2, Src-homology-2 domain; Shc, Src homology 2 domain-containing transforming protein; wt, wild-type n Corresponding author. E-mail address: [email protected] (L. Raptis). 1 Present address: Department of Chemical and Physical Sciences, University of Toronto, Mississauga, Ontario, Canada L5L 1C6. 2 Present address: Faculty of Law, University of Toronto, Toronto, Ontario, Canada M5S 2C5. 3 The second and third authors contributed equally to this work. http://dx.doi.org/10.1016/j.yexcr.2015.07.013 0014-4827/& 2015 Elsevier Inc. All rights reserved.

aggregation of two hemichannels, each containing six connexins, contributed by each neighboring cell. Results from a number of labs have indicated that gap junctional, intercellular communication (GJIC) is interrupted in cells transformed by oncogenes such as vSrc [42], vRas [3,8] and others [25], while a number of tumor lines and primary tumor cells have reduced GJIC [21,57]. The study of the middle Tumor antigen of polyoma virus (mT) has shed light on a number of important concepts of signal transduction. mT is a membrane-bound, 421-amino-acid protein which is essential for transformation of cultured rodent fibroblasts by this virus. In fact, targeting mT expression specifically to mouse mammary epithelial cells results in the induction of multifocal mammary tumors with 100% penetrance (reviewed in [38]). mT initially associates with protein phosphatase 2A and this allows the recruitment of members of the cellular Src protooncogene product family of tyrosine kinases (cSrc, cYes and Fyn, reviewed in [16]). This association stimulates their enzymatic activity to phosphorylate a number of substrates including mT itself and cSrc, and in this manner mT offers a scaffolding platform for the assembly of cellular signaling proteins [16,28]. Thus, mT becomes

224

M. Geletu et al. / Experimental Cell Research 336 (2015) 223–231

phosphorylated on three major tyrosine residues, which provide docking sites for the PTB (phosphotyrosine binding) domain of Shc (Src homology 2 domain-containing transforming protein, p-tyr250 [11,17]), the SH2 (Src homology 2) domain of p85, the adaptor subunit of class IA phosphatidylinositol 3-kinase (PI3k, p-tyr315 [13,32,55,59]), and PLC-γ (phospholipase Cγ, p-tyr322 [54]), resulting in the activation of their respective downstream signaling cascades. As a result, point mT mutations can disrupt its ability to activate each pathway specifically. Later examination of the amino-acid sequence surrounding p-tyr250 (NPT250YSVM) indicated that it contains a secondary PI3k binding site, so that the tyr250-phe mutation affects both the Shc/Ras/Erk, and to some degree the PI3k/Akt pathway. In addition, tyr315 is a secondary binding site for an unidentified protein which is critical for transformation [27]. Early data demonstrated that wild-type (wt) mT expression can effectively suppress GJIC in mouse fibroblasts [6,43]. However, the functions required for mT-mediated, gap junction closure have not been identified. Given the divergent roles of the different pathways induced by the mT tyrosine phosphorylation sites, we set out to examine the effect of different mT mutants upon GJIC. We chose the rat liver epithelial cell line T51B which normally has extensive GJIC, and mT mutants were stably expressed through transfection or retroviral infection. GJIC was examined using a novel technique of in situ electroporation we developed: Cells were grown on a glass slide, part of which was coated with electrically conductive, optically transparent, indium-tin oxide. An electric pulse was applied in the presence of the fluorescent dye, Lucifer yellow (LY), causing its penetration into the cells growing on the conductive part of the slide, and the migration of the dye through gap junctions to the non-electroporated cells growing on the non-conductive area was microscopically observed under fluorescence illumination [23,47]. The results show that, although even low levels of wt-mT are sufficient to interrupt gap junctional, intercellular communication, GJIC suppression requires binding to Shc, that is activation of the Ras pathway. Interestingly, PI3k activation is not required for GJIC suppression by mT. In the contrary, expression of activated forms of the catalytic p110 subunit of PI3k actually increase GJIC, indicating that PI3k has a positive effect upon gap junctional communication in this system.

2. Materials and methods 2.1. Cell lines and gene expression The T51B, rat liver epithelial line and derivatives expressing NG59 and 248H have been described previously [22,48]. All cells were grown in DMEM with 10% fetal calf serum. Extra care was taken to ensure that cell seeding was uniform, by passing cells at subconfluence, when cell to cell adhesion was low. The mT mutant retroviral plasmids expressing 250F-mT and 315F-mT (in a pWZLBlast backbone) were a generous gift of Drs. Thomas Roberts and Tamara Utermark, Harvard Medical School. Plasmids were transfected into the Phoenix ecotropic packaging line and the culture supernatant containing the viral vector was used to infect T51B cells which were then selected with 2.5 μg/mL blasticidin (BioShop, Cat. number BLA477.100). Individual clones were picked, expanded into lines, frozen and used in further experimentation. The myristylated, constitutively active p110 subunit, retroviral vector plasmid and its kinase-dead counterpart were from Addgene; Cat. Number 10836 (pLNCX Myr His p110), and Cat. Number 10837 (pLNCX Myr His p110 KD), respectively. Following transfection in the Phoenix ecotropic packaging line, the culture supernatant was used to infect T51B cells which were selected with 300 μg/mL G418 (BioShop, Cat. Number GEN418). Individual

clones were picked, expanded into lines, frozen and used in experiments. The p110 constitutively active mutants E545K and H1047R, in a pBabe-puro background cloned at the SalI to BamH1 sites (Addgene #2525 and #12524, respectively), were a generous gift from Dr. T. Roberts. T51B, rat liver epithelial cells were infected with the culture supernatant from transfected Phoenix ecotropic cells, selected for puromycin resistance and individual clones used in experiments. Mutationally activated RasL61 was expressed through infection with the culture supernatant of the packaging line Nψ6 which produces recombinant retrovirus carrying a L61activated, c-Ha-Ras-1 cDNA in the pDOL retroviral vector [44], a gift of Dr. T. Roberts, Harvard Medical School. The PI3k inhibitors, LY294002 and Wortmannin, were purchased from Sigma. Cells were treated with 20 μM LY294002 (Cat. Number L9908) for 20 h or with 100 nM Wortmannin (Cat. Number W1628) for 20 h. For Western blotting analysis lysates from cells that were serum-starved for 24 h were prepared by RIPA (0.1% sodium dodecyl sulfate, 1% Nonidet-P40, 0.5% sodium deoxycholate, 50 mM Tris pH 8.0, 150 mM NaCl) extraction. This was especially important for Cx43 analysis. Following a careful protein determination using the Bicinchoninic Acid assay (Sigma, Cat. Number B9643), 20 μg total protein were resolved by polyacrylamide gel electrophoresis. Blots were subsequently cut into strips and probed with antibodies against: Cx43 (a gift of Dr. Kardami, University of Manitoba); AktpS473 (Cell Signaling, rabbit D9E, Cat. Number 4060X, lot 5, used at 1:1,000); p110α (Cell Signaling, rabbit C73F8, Cat. Number 4249S, lot 3, used at 1:1,000); Ras (Cell Signaling, rabbit mAb 27H5, Cat. Number 3339, lot 3, used at 1:1,000); pErk1/2 (a gift of Dr. Eric Schaefer of QCB [45]); mT (F4, the antibody producer line was a gift from Drs. P. Bourgaux and L. Delbecchi, University of Sherbrooke); and cleaved PARP (Cell Signaling, Cat. Number 9548, lot 4, used at 1:5,000). Antibodies against β-actin (Cell Signaling, Cat. Number 3700S, lot 8, used at 1:20,000); tubulin (Cell Signaling, Cat. Number 2125, used at 1:10,000); or Hsp90 (Thermo Scientific, Cat. Number PA5-17402, used at 1:10,000) were used as loading controls [26]. 2.2. Examination of gap junctional communication The technique described before [2,21,23,47] was used. Briefly, cells were grown on a glass slide which was partially coated with electrically conductive and transparent Indium–Tin oxide. An electrical pulse of precisely controlled intensity and duration was applied in the presence of the fluorescent dye, Lucifer yellow (LY, Biotium, potassium salt, Cat. Number 80016, lot 010128, 5 mg/ml in DMEM without Calcium). The pulse opens transient pores on the cell membrane through which Lucifer yellow enters and the migration of the dye through gap junctions to cells on the neighboring, non-electroporated side was microscopically observed under fluorescence illumination. All experiments were performed at least three times, where transfer from more than 200 cells was calculated. To avoid subjective bias, samples were coded before experiments. The equipment (In situ Porator) was supplied by Cell Projects Ltd. UK.

3. Results 3.1. Src binding and tyr250 phosphorylation are required for GJIC suppression by mT Results from a number of labs have indicated that binding of Src family kinases is important for mT-mediated transformation [28]. Regarding gap junctional communication, we and others previously demonstrated that mT can, in fact, suppress GJIC [6,43].

M. Geletu et al. / Experimental Cell Research 336 (2015) 223–231

T51B

mutants expressed

t-m T 31 5F 24 8H N G 59 25 0F dl 23

cell line

w

-

T51B mT

48 2

100

32

83

93

96

23

1

2

3

4

5

6

7

-

N

band intensity (%)

225

100

63 band intensity (%)

t-m T 25 0F 31 5F dl 23

8H

w

24

59 G

mutants expressed

Hsp90

Akt pser473 10

15

100

45

81

18

wt-mT

17

100

2

3

-

5

4

6

7

10

23 dl

H

5F

8H

31

w

t-m

T

mutants expressed

47

R

1

24

Hsp90

48 band intensity (%)

p44 p42 pErk1/2 80

29

28

100

41

49

100

band intensity (%)

59 G N

23

6

dl

8H

5

24

0F

-

4

25

t-m w 48

5F

3

T

2

31

Hsp90 1

mutants expressed

315F

Cx43 28

40

17

75

100

12

248H

40

100

Hsp90 1

2

3

4

5

6

7

Fig. 1. A: Expression of wt-mT antigen and its mutants in T51B cells. Lysates from the parental T51B or derivative lines expressing wt-mT or the mutants indicated, were resolved by SDS electrophoresis and probed for mT or Hsp90 as a loading control (see Section 2). Numbers under the lanes refer to values obtained by imaging analysis, with the value of wt-mT taken as 100%. Numbers at the left refer to Molecular weight markers. B: Akt-p473 levels in T51B cells and derivatives expressing wt-mT or its mutants. Lysates from T51B cells or derivatives expressing wt-mT or the mutants indicated were resolved by SDS-gel electrophoresis and probed for Akt-p473, or Hsp90 as a loading control (see Section 2). Numbers under the lanes refer to values obtained by imaging analysis, with the highest value, of 248H-mT, taken as 100%. Numbers at the left refer to Molecular weight markers. C: Erk1/2 levels in T51B cells and derivatives expressing wt-mT, its mutants or the H1047R, p110 mutant. Lysates from T51B cells or derivatives expressing wt-mT or the mutants indicated, or the activated p110 mutant H1047R were resolved by SDSgel electrophoresis and probed for p-Erk1/2, or Hsp90 as a loading control (see Section 2). Numbers under the lanes refer to values obtained by imaging analysis, with the highest value, of 315F-mT, taken as 100%. Numbers at the left refer to Molecular weight markers. D: Cx43 levels in T51B cells before or after expression of mT or its mutants. Lysates from T51B cells or derivative lines stably expressing wtmT or the mutants indicated were resolved by SDS-gel electrophoresis and probed for Cx43 or Hsp90 as a loading control (see Section 2). Numbers under the lanes refer to values obtained by imaging analysis, with the highest value, of 248H-mT, taken as 100%. Numbers at the left refer to Molecular weight markers.

Fig. 2. GJIC in T51B cells and derivatives expressing wt or mutant mT. T51B (a and b), T51B-mT (c and d), T51B-315F (e and f) or T51B-248H (g and h) cells were plated in electroporation chambers and subjected to a pulse in the presence of Lucifer yellow. After washing away the unincorporated dye, cells from the same field were photographed under fluorescence (b, d, f and h) or phase contrast (a, c, e and g) illumination. Cells at the edge of the electroporated area (arrow) which are loaded with LY through electrical pulse application are marked with a star, and cells at the non-electroporated area which received LY through gap junctions are marked with a dot [47]. Magnification: 240  . Note the extensive gap junctional communication of T51B-248H cells (g and h).

Moreover, even low levels of mT,
14% the levels required for anchorage-independent proliferation and tumorigenicity of ratF111 fibroblasts, were found to be sufficient to block GJIC in rat and mouse fibroblasts [43], indicating an exquisite sensitivity of GJIC to very low wt-mT levels. To examine the mT functions required for GJIC suppression in T51B cells we made use of point mutants where specific tyrosine residues were converted to phenylalanine ([27], see Section 2). Following stable expression of each mutant (Fig. 1A), cells were plated in electroporation chambers and an electrical pulse was applied in the presence of LY. GJIC was quantitated by assessing the number of cells into which LY had penetrated by diffusion through gap junctions, per cell where LY entered by electroporation (see Section 2). As shown in Fig. 2 (c

226

Table 1 A.Effect of mT and its mutants upon GJIC. Cell line

250a

315a

pathways activatedb

T51B wt-mT NG59 250F 248H 315F dl23

 NPTYSVM N/A NPTFSVM NHTYSVM NPTYSVM NPTYSVM

 YMPM N/A YMPM YMPM FMPM 

cSrcc  þ  þ þ þ þ

GJICf Shcd  þ    þ þ

PI3ke  þ   /þ þþ  

3.0 7 0.9 0.2 7 0.1 3.17 1.2 3.5 7 1.0 5.05 7 1.1 0.2 7 0.1 0.17 0.1

Cell line

treatment

Akt-p473g (%)

GJIC

T51B T51B T51B T51B-myr p110 T51B-myr p110KD T51B-E545K T51B-H1047R T51B-myr p110-RasL61

 Wortmannin LY294002     

1 70.2 0.20 70.1 0.25 70.1 14.3 73.2 1 70.1 13 72.1 12 73.4 14.3 73.2

3.0 7 0.9 0.2 7 0.1 0.3 7 0.1 6.17 1.4 3.0 7 1.5 5.8 7 1.5 5.17 1.1 0.4 7 0.2

a

250, 315: Status of aminoacids at the main Shc (Y250) or PI3k (Y315) binding sites. Pathways activated by wt-mT or its mutants [27]. c cSrc: Stimulation of the cSrc kinase activity. d Shc: Activation of the Shc/Grb2/Ras/Raf/Mek/Erk pathway. e PI3k: Activation of the PI3k/Akt pathway. N/A: Not applicable. f GJIC was assessed by in situ electroporation (see Materials and Methods). Quantitation was achieved by dividing the number of cells into which the dye had transferred through gap junctions (denoted by dots, Figs. 2, 3), by the number of cells at the edge of the electroporated area (denoted by stars). Numbers are averages 7 SEM of at least three experiments, where transfer from a minimum of 200 cells was examined. g Akt-p473 levels were examined by Western blotting in the parental T51B, and cells stably expressing myr-p110 or a kinase-dead mutant, or the constitutively active, E545K or H1047R, p110 mutants, or treated with Wortmanin or LY294002 (see Materials and Methods). Numbers refer to relative amounts following densitometric scanning, with the levels in the parental T51B taken as 1. Averages of at least 2 experiments from at least 5 independent clones expressing myr-p110, E545K or H1047R 7 SEM are presented. b

M. Geletu et al. / Experimental Cell Research 336 (2015) 223–231

B. Effect of PI3k upon GJIC

M. Geletu et al. / Experimental Cell Research 336 (2015) 223–231

and d vs a and b), T51B cells expressing wt-mT had essentially no GJIC, while GJIC in cells expressing the NG59 mutant-mT which does not bind Src, is indistinguishable from the parental T51B (Table 1), indicating that binding to Src family kinases (perhaps in combination with the PP2A adaptor) is likely to be required for the mT-triggered, GJIC suppression. The site encompassing tyr250 (NPTY250) is the main site of binding of the PTB domain of Shc onto mT. In fact, a tyrosine 250 to phenylalanine mutation (250F-mT) was found to completely abolish Shc association with mT [11,18,39]. To assess the importance of tyr250 phosphorylation, GJIC was examined in T51B cells expressing 250F-mT through retroviral transduction (Fig. 1A, see Section 2). As shown in Table 1, 250F-mT expressing, T51B cells were found to have similar GJIC values as the parental T51B. These results indicate that the 250F-mT mutant is unable to suppress GJIC, that is tyr250 phosphorylation is, in fact, required for GJIC suppression by mT. 3.2. Tyr315 phosphorylation is not required for GJIC suppression by mT The phosphorylated tyrosine residue 315 (mT-315YMPM motif) functions as a binding site for an SH2 domain of the p85 regulatory subunit of the PI3-kinase [51,60]. Replacing tyr315 with phenylalanine (315F-mT) was found to abrogate p85 association and PI3k activation by mT [31] and to reduce mT's transforming ability in a variety of systems including T51B cells [48]. To examine the involvement of the tyr315 site, the 315F-mT mutant was expressed in T51B cells. Levels of the pErk1/2 Ras effector and the ser473phosphorylated form of the Ser/thr kinase Akt, PI3k effector, which correlates with Akt activity [20], as well as GJIC levels were subsequently assessed as above (Fig. 1C and B). As expected, 315F-mT expressing cells had similar Akt-p473 levels as the parental T51B (Fig. 1B, lanes 1 vs 6). Interestingly however, 315F-mT mutant expression essentially eliminated gap junctional communication in a manner indistinguishable from wt-mT (Fig. 2e and f), indicating that, despite its importance in transformation, tyr315 is not required for GJIC suppression. In addition, expression of mutant dl23, which carries a deletion of 34 aminoacids encompassing tyr315 and tyr322 (binding site for the SH2 domain of PLCγ) [49] induced a complete GJIC suppression, in a manner similar to 315FmT (Table 1). Taken together, the above data indicate that activation of the Shc/Grb2/Ras pathway, perhaps in conjunction with other pathways emanating from mT such as 14.3.3 (ser257-mT [12]), is sufficient to drive the dramatic reduction in gap junctional communication, in the absence of full activation of the PI3k/Akt or the PLCγ (which would activate PKC, a known GJIC suppressor [37]) pathways and neoplastic transformation. 3.3. Shc binding is required for GJIC suppression by mT A detailed examination of the sequences surrounding the Shc binding site (NPT250YSVM) through mutation of the aminoacids following 250Y to alanine (NPT250YAAA) previously revealed that 250 Y functions as a secondary binding site for PI3 kinase [27]. As a result, the 250F-mT mutant, besides being unable to bind Shc, it is also partially defective in binding of the p85 subunit, and activation of PI3k. Therefore, to examine the role of Shc specifically, upon the mT-induced GJIC suppression, we made use of the P248H mutant. In this mutant, binding of Shc is impaired, although the secondary PI3k binding site (250YSVM) is intact [18]. Possibly due to the fact that PI3k is free to bind Y250, without competition from Shc, the 248H mutant was found to activate PI3k/Akt more than wt-mT [27]. To verify the effect of the P248H mutation upon Aktp473 and GJIC, this mutant was expressed in T51B cells and Aktp473 levels examined as above. As shown in Fig. 1B, expression of

227

248H-mT caused a greater increase in Akt-p473 levels than wt-mT (lane 3 vs 4). Interestingly, the results indicated that T51B-248HmT cells had distinctly higher GJIC than the parental line (Fig. 2g and h, and Table 1), pointing to the exciting possibility that PI3 kinase activation by mT may be actually increasing gap junctional communication even in the face of PLCγ/PKC activation, as long as the GJIC-suppressing, Ras/Erk pathway is not activated. 3.4. PI3k plays a positive role upon GJIC The PI3 kinase holoenzyme consists of two parts: A catalytic, 110 kDa (p110) and a regulatory, 85 kDa subunit. Activation of membrane tyrosine kinases, or expression of oncogenes such as mT or Src recruits PI3 kinase to the membrane through the three SH2 (Src-homology-2) domains of p85 which bind phosphorylated, pYXXM motifs. There PI3k initiates signaling cascades, by generating second messenger phosphatidylinositol-3,4,5-trisphosphate [PtIns(3,4,5)P3], which provides a docking site for signaling molecules with plekstrin-homology domains (reviewed in [35]). Akt is the most important PI3k effector and is recruited to the plasma membrane through PtIns(3,4,5)P3 binding and activated by phosphorylation at S473 by mTORC2 (mammalian target of rapamycin complex-2) and at T308 by the phosphatidylinositoldependent kinase 1. Akt plays a pivotal role in cellular survival and metabolism, in addition to cell proliferation [36]. In fact, stable expression of a myristylated form of the catalytic subunit of PI3k, p110α which renders it constitutively active in murine fibroblasts (myr-p110), was shown to be able to induce morphological transformation in mouse NIH3T3 fibroblasts [4]. To examine whether PI3k alone is sufficient to increase GJIC, myr-p110 was expressed through retroviral transduction in T51B cells (see Materials and Methods). As shown in Fig. 3B, myr-p110 expression caused a dramatic increase in Akt-p473 levels, compared to the control, kinase-dead mutant (lanes 1 vs 2). Interestingly, myr-p110 expressing T51B cells had higher GJIC than the parental T51B (Fig. 3C, a and b vs c and d), while the myr-p110, kinase-dead vector had no effect upon GJIC (Table 1), indicating that myr-p110 expression does in fact increase GJIC. p110 is found to be mutated in a large variety of cancers. Two mutations in particular, E545K and H1047R stimulate kinase activity and exert a strong oncogenic drive [35]. To further examine the role of PI3k upon GJIC, these mutants were expressed in T51B cells through retroviral transduction and gap junctional communication examined as above. As shown in Table 1, expression of the E545K or H1047R mutants increased GJIC, further indicating that PI3 kinase plays a positive role upon GJIC. To further confirm this conclusion, the effect of PI3k inhibition upon GJIC was examined next. T51B cells were plated in electroporation chambers and treated with the PI3k inhibitors LY294002 or Wortmannin for 24 h, prior to electroporation of Lucifer yellow (see Section 2). The results demonstrated a dramatic GJIC reduction (Table 1), indicating that PI3k activity is, in fact, required for gap junctional communication in T51B cells. Finally, to examine whether activated Ras can disrupt GJIC in the face of high PI3k/Akt activity, mutationally activated, RasL61 was expressed with a retroviral vector in myr-p110 expressing, T51B cells. As shown in Fig. 3C (e and f), RasL61 expression in myr-p110 cells abolished GJIC, indicating that the force of RasL61 to suppress GJIC prevails.

4. Discussion Results from a number of labs have demonstrated the importance of activation of both the Ras and PI3k pathways in transformation of rodent fibroblasts by mT, as assessed by agar growth, foci formation or tumorigenicity [58]. GJIC suppression is

M. Geletu et al. / Experimental Cell Research 336 (2015) 223–231

11 0

p110 69

β-actin 2

3

4

5

6

R 47

K

10

45

H

E5

yr m

16

93

100

93

44

79

100

81

1

2

3

4

48

100

48

1

m

band intensity (%)

yr

95

yr

21

100

Akt pser473

60

-p

11 0

KD

mutants expressed

m

-

m

-p

R 47

70

10

45

16

H

-

E5

band intensity (%)

K

mutants expressed

yr

-p

T51B

cell line

-p 11 0

11 0

KD

228

Cx43

100

Hsp90

T51B

myr-p110

myr-p110 Ras

R

-

cell line

as

T51B myr-p110

p21 Ras

20

α-tubulin

48

1

2

Fig. 3. A: p110 levels in T51B cells and derivatives transduced for myr-p110, E545K or H1047R. Lysates from T51B cells (lanes 1 and 4), or clones expressing E545K (lane 2), H1047R (lane 3), myr-p110 (lane 5) or a control clone expressing a kinase-dead myr-p110 (lane 6), were resolved by electrophoresis and probed for p110 (top panel), or βactin as a loading control (bottom panel) (see Section 2). Numbers at the left refer to Molecular weight markers. B: Akt-pser473 and Cx43 levels in T51B cells expressing myrp110, E545K or H1047R. Lysates from T51B cell derivatives expressing kinase dead, myr-p110 (lane 1), myr-p110 (lane 2), E545K (lane 3), or H1047R (lane 4), were resolved by SDS-gel electrophoresis and probed for Akt-pser473, Cx43 or Hsp90 as a loading control (see Section 2). Numbers under the lanes refer to Cx43 values obtained by imaging analysis, with the highest value, of H1047R, taken as 100%. Numbers at the left refer to Molecular weight markers. C: myr-p110 increases GJIC in T51B cells. T51B (a and b), T51B-myr-p110 (c and d) or T51B-myr-p110-RasL61 (e and f) cells were plated in electroporation chambers and subjected to a pulse in the presence of Lucifer yellow. Following washing of the unincorporated dye, cells from the same field were photographed under fluorescence (b, d and f) or phase contrast (a, c and e) illumination. Cells at the edge of the conductive area which were loaded with LY through electroporation were marked with a star and cells at the non-electroporated area which received LY through gap junctions were marked with a dot as in Fig. 2. Arrows point to the edge of the electroporated area. Magnification: 240  . Note the extensive gap junctional communication of T51B-myr-p110 cells (a and b). Right panel: Lysates from T51B-myr-p110 cells, before (lane 1) or after (lane 2) RasL61 expression were probed for Ras or αtubulin as a control, as indicated.

one of the parameters invariably associated with neoplasia [21,24]. The mechanism of GJIC disruption by activated Src has been extensively studied (reviewed in [42]), and Src is known to phosphorylate Cx43 directly, at tyr247 and tyr265. However, whether it is the direct Cx43 phosphorylation by Src that inhibits GJIC, or whether its effector pathways PLCγ/PKC and Erk1/2 which are known to suppress GJIC in their own right play a more important role, is still controversial ([33,34,40,50,61], reviewed in [5]). In this communication we demonstrate the effect of specific pathways in the context of mT-mediated transformation, using a technique for GJIC examination which makes the quantitation of GJIC in a large number of cells possible. 4.1. The Ras pathway is important for GJIC suppression by mT Results from a number of labs previously demonstrated that constitutively active Ras is neoplastically transforming on its own,

while binding of Shc and activation of the Ras/Raf/Erk pathway is important for transformation by the mT:pp60c-Src complex [1,28,46]. Regarding GJIC, it was previously demonstrated that mutationally activated Ras suppresses GJIC [8,10]. In addition, inhibition of Ras in rat fibroblasts transformed by mutationally activated Src reinstated gap junctional communication [30]. Conversely, mT expression (hence cSrc activation) in Ras-deficient fibroblasts was unable to eliminate GJIC [8–10]. These data underline the importance of the Ras pathway in GJIC suppression by Src, activated either through a mutation or mT binding. Our present findings further indicate that mutations 250F or 248H, either of which eliminates Shc binding and Ras activation by mT, impair mT's ability to suppress GJIC. Conversely, the 315F-mT mutant which is able to activate Ras suppresses GJIC to an extent similar to wt-mT. Taken together, these results reinforce and extend previous observations. It is especially remarkable that the dl23-mT mutant (which has a deletion of 34 aminoacids including tyr315 and

M. Geletu et al. / Experimental Cell Research 336 (2015) 223–231

tyr322), although it was expressed at lower levels than any of the others (Fig. 1A), was able to eliminate GJIC (Table 1). This finding indicates that the Ras pathway is exquisitely capable of driving gap junction closure without a contribution from PLCγ/PKC (tyr322mT), which is known to suppress GJIC through Cx43 phosphorylation at S368 and S372. In addition, since the 248H-mT mutant does activate the cSrc kinase as much as wt-mT [48], the above findings also show that activation of cSrc by mT, hence presumably direct Cx43 phosphorylation by the mT:pp60c-Src complex is not sufficient for GJIC suppression in this system. 4.2. PI3k plays a positive role upon gap junctional communication Activation of PI3k by mT is necessary for full neoplastic transformation [27,55]. Our results however, showing that 250F-mT which does allow at least partial PI3k activation, is defective in GJIC suppression, indicate that PI3k does not transmit mT signals to gap junction closure. The fact that mutant 248H-mT which, unlike 250F-mT (which is partially deficient in PI3k activation as well [27]) activates PI3k/Akt more than wt-mT (Fig. 1B, lane 3 vs 4), has actually higher GJIC than the parental T51B cells (Fig. 2g and h vs a and b), further reinforces this conclusion. Therefore, PI3k activation can, in fact, overcome a potentially GJIC suppressive effect of PLCγ/PKC (Y322-mT), as long as the Ras pathway is not activated. In the presence of high Ras activity however, as upon wt-mT expression, Ras's GJIC-suppressive effect prevails, with gap junction closure as a result. This holds true for myr-p110 expressing cells as well, where RasL61 expression suppressed GJIC, although the exact degree might depend on levels of myr-p110 and RasL61 expression. The fact that PI3k activation by mT increases GJIC prompted us to explore a potential positive role of constitutively active PI3k upon GJIC. Interestingly, retroviral expression of PI3k forms, activated through myristylation (Fig. 3C) or mutation (Table 1), increased GJIC. Conversely, PI3k pharmacological inhibition in T51B cells abolished GJIC (Table 1). Taken together, these data indicate that PI3k does in fact play a positive role in the maintenance of gap junction function in this system (Fig. 4). Despite extensive efforts, the effect of the PI3k/Akt pathway

cSrc

•

250

Shc

mT

•

315

p85 p110

Ras

GJIC

Akt

GJIC

Fig. 4. Model of the effect of RasvsPI3k activation by mT upon GJIC mT binds to and is phosphorylated by tyrosine kinases of the cSrc family. The mT-PY250 site is a binding site for the Shc adaptor, which triggers the activation of the Ras/Erk pathway and GJIC suppression. On the other hand, mT-PY315 is the binding site for the 85 kDa regulatory subunit of the PI3 kinase, which triggers activation of Akt and an increase in GJIC. Upon wt-mT expression the Ras pathway prevails, with gap junction closure as a result. At the same time, both pathways are required for full neoplastic conversion by mT.

229

upon GJIC is unclear. Previous results indicated that Akt-mediated phosphorylation stabilises membrane-localized Cx43 [19,52,53], and that Akt is required for the maintenance of steady-state Cx43 levels and GJIC in osteoblasts [7]. However, it was also shown that Akt1 (but not Akt2 or Akt3) is actually positively involved in the disruption of gap junctions caused by activated Src in rat and mouse fibroblasts [29], while PI3k was found to have both positive and negative effects upon GJIC in the Xenopus oocyte system [40]. Our data, showing that pharmacological inhibition of PI3k abolished GJIC in T51B cells (Table 1), while PI3 kinase, activated by mutant-mT, membrane translocation (myr-p110) or activating mutations (E545K and H1047R) increases GJIC, indicate that PI3k is required for gap junctional communication, and that in an activated form it can increase GJIC in rat liver epithelial cells. Barring differences due to the particular cell types used, it is possible that PI3k is activating all three Akt isoforms in T51B cells, so that the net effect is a GJIC increase. In any event, our data are consistent with previous findings where Akt was found to be responsible for a GJIC increase at the initial stages following acute Src activation or growth factor stimulation [50]. However, our results clearly indicate that chronic Src activation (through mT expression) requires the Ras/Erk pathway for GJIC suppression, while PI3k/Akt actually increases GJIC. The exact mechanism is under investigation. Although it is sufficient to dramatically suppress GJIC, activation of the Ras or PI3k pathway alone is not sufficient for full, mTinduced neoplastic conversion of mouse or rat fibroblasts [27,28]. Regarding T51B cells, the 248H-mT mutant which activates PI3k/ Akt but not Shc/Ras did induce anchorage independence and cells were able to grow into tumors in syngeneic rats, but the tumors had a different histology than wt-mT transformed T51B. On the other hand, activation of the Ras pathway alone with mutants 315F-mT or dl23-mT is not sufficient for anchorage independence or tumorigenicity [48]. These findings indicate that the interruption of gap junctional communication is independent of full neoplastic conversion and the concomitant changes in cell shape in this system. It was previously demonstrated that a mutation in the main PI3K binding site of mT, 315YMPM to 315YAAA abrogated PI3K activation, although this mutant could still transform mouse Balb/ c3T3 fibroblasts to approximately 73% of the wt, as measured by foci formation on plastic or colonies in soft agar [27]. This was taken as an indication that another protein (termed protein X) whose binding site centers around 315Y, is necessary for mT transformation. However, since 248H-mT which has an intact 315Y site displayed high GJIC, although protein X should have been able to bind to 315Y, it appears that the protein X cannot be suppressing GJIC on its own in this system. Akt promotes cellular survival through multiple mechanisms (reviewed in [36]). Akt blocks the function of pro-apoptotic proteins such as BAD, which is a direct phosphorylation target of Akt [14,15]. Akt also phosphorylates the E3 ubiquitin ligase MDM2, and this promotes MDM2 translocation to the nucleus where it triggers degradation of the p53 tumor suppressor. Global induction of apoptosis with etoposide, cycloheximide or puromycin was shown to lead to a loss of cell coupling, probably due to caspase-3mediated degradation of Cx43, in primary bovine lens epithelial and mouse NIH3T3 fibroblasts [56]. Therefore, apoptosis inhibition due to PI3k activation through expression of 248H-mT or the p110 constructs, would increase Cx43 and gap junction function. In fact, cells expressing wt-mT displayed higher levels of cleaved PARP than cells expressing 248H-mT or the p110 constructs, before or after LY294002 treatment (Fig. S3, Supplementary data), as previously demonstrated in other systems. However it is also possible that PI3k may be affecting Cx43 mRNA levels, as observed before in osteoblasts [7]. A modest reduction in total Cx43 protein levels was noted in wt-mT-expressing cells, and an increase upon 248H-

230

M. Geletu et al. / Experimental Cell Research 336 (2015) 223–231

mT (Fig. 1D), myr-p110, E545K or H1047R (Fig. 3B) expression, but Cx43 phosphorylation at specific sites may also play a very important role [5]. The precise mechanism is under investigation.

5. Conclusions Our results demonstrate that, although both the Ras and PI3k pathways are required for transformation, PI3k is not transmitting signals from the mT:cSrc complex leading to gap junction closure in T51B cells. In the contrary, PI3k actually promotes gap junctional communication despite the fact that when activated, either by mT, mutation, or translocation to the membrane, PI3k can act as an oncogene. Most importantly, these results demonstrate a dramatic difference in the response of gap junctional communication to the two pathways, which emanate from a multitude of oncogenes including the Src family and growth factor receptors, and are jointly required for full neoplastic conversion. Coupled with our recent findings indicating that activated Stat3, an oncogene with a potent survival action, is also required for gap junction function [21,22], it is tempting to speculate that neoplastic transformation requires PI3k/Akt and survival, as well as Ras/Erk, while a survival function is actually required for the maintenance, rather than suppression, of gap junctional communication. In wt-mT, where both the Ras and PI3k pathways are activated the Ras pathway prevails, hence the net effect is gap junction closure. This novel role of Src/PI3k may be an important regulatory step in the progression of tumors that exploit such a pathway.

Conflicts of interest The authors have no conflict of interest to declare.

Acknowledgments The authors thank Drs. Thomas Roberts and Tamara Utermark (Harvard Medical School) for a gift of the 248H and 315F mT mutants, the Nψ6 line and the E545K and H1047R, p110 mutants; Dr. Normand Marceau of Laval University for cell lines; Drs. Pierre Bourgaux and Louis Delbecchi (University of Sherbrooke) for a gift of the F4 mT antibody producer line; Dr. Elissavet Kardami (University of Manitoba, Canada) for a generous gift of connexin-43 antibody and Kevin Firth, P.Eng., of Ask Sciences Products, Kingston, Ontario for engineering design. The financial assistance of the Natural Sciences and Engineering Research Council of Canada (NSERC), the Canadian Institutes of Health Research (CIHR), the Canadian Breast Cancer Foundation (Ontario Chapter), the Canadian Breast Cancer Research Alliance, the Ontario Centers of Excellence, the Breast Cancer Action Kingston and the Clare Nelson bequest fund through grants to LR is gratefully acknowledged. S. Guy was supported by an NSERC studentship. MG was supported by postdoctoral fellowships from the US Army Breast Cancer Program, the Ministry of Research and Innovation of the Province of Ontario and the Advisory Research Committee of Queen's University.

Appendix A. Supplementary Information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.yexcr.2015.07.013.

References [1] A. Aleshin, R.S. Finn, SRC: a century of science brought to the clinic, Neoplasia 12 (2010) 599–607. [2] A. Anagnostopoulou, J. Cao, A. Vultur, K.L. Firth, L. Raptis, Examination of gap junctional, intercellular communication by in situ electroporation on two coplanar indium-tin oxide electrodes, Mol. Oncol. 1 (2007) 226–231. [3] M.M. Atkinson, J.D. Sheridan, Altered junctional permeability between cells transformed by v-ras, v-mos, or v-src, Am. J. Physiol. 255 (1988) C674–C683. [4] K.R. Auger, J. Wang, R.P. Narsimhan, T. Holcombe, T.M. Roberts, Constitutive cellular expression of PI 3-kinase is distinct from transient expression, Biochem. Biophys. Res. Commun. 272 (2000) 822–829. [5] L.N. Axelsen, K. Calloe, N.H. Holstein-Rathlou, M.S. Nielsen, Managing the complexity of communication: regulation of gap junctions by post-translational modification, Front. Pharmacol. 4 (2013) 130. [6] R. Azarnia, W.R. Loewenstein, Polyomavirus middle t antigen downregulates junctional cell-to-cell communication, Mol. Cell. Biol. 7 (1987) 946–950. [7] R. Bhattacharjee, M. Kaneda, K. Nakahama, I. Morita, The steady-state expression of connexin43 is maintained by the PI3K/Akt in osteoblasts, Biochem. Biophys. Res. Commun. 382 (2009) 440–444. [8] H.L. Brownell, R. Narsimhan, M.J. Corbley, V.M. Mann, J.F. Whitfield, L. Raptis, Ras is involved in gap junction closure in mouse fibroblasts or preadipocytes but not in differentiated adipocytes, DNA Cell Biol. 15 (1996) 443–451. [9] H.L. Brownell, J.F. Whitfield, L. Raptis, Cellular Ras partly mediates gap junction closure by the polyoma virus middle Tumor antigen, Cancer Lett. 103 (1996) 99–106. [10] H.L. Brownell, J.F. Whitfield, L. Raptis, Elimination of intercellular junctional communication requires lower Rasleu61 levels than stimulation of anchorageindependent proliferation, Cancer Detect. Prev. 21 (1997) 289–294. [11] K.S. Campbell, E. Ogris, B. Burke, W. Su, K.R. Auger, B.J. Druker, B. S. Schaffhausen, T.M. Roberts, D.C. Pallas, Polyoma middle tumor antigen interacts with SHC protein via the NPTY (Asn–Pro–Thr–Tyr) motif in middle tumor antigen, Proc. Natl. Acad. Sci. USA 91 (1994) 6344–6348. [12] J. Cheng, J.A. DeCaprio, M.M. Fluck, B.S. Schaffhausen, Cellular transformation by Simian Virus 40 and Murine Polyoma Virus T antigens, Semin. Cancer Biol. 19 (2009) 218–228. [13] S.A. Courtneidge, A. Heber, An 81 kd protein complexed with middle T antigen and pp60c-src: a possible phosphatidylinositol kinase, Cell 50 (1987) 1031–1037. [14] S.R. Datta, H. Dudek, X. Tao, S. Masters, H. Fu, Y. Gotoh, M.E. Greenberg, Akt phosphorylation of BAD couples survival signals to the cell-intrinsic death machinery, Cell 91 (1997) 231–241. [15] S.R. Datta, A.M. Ranger, M.Z. Lin, J.F. Sturgill, Y.C. Ma, C.W. Cowan, P. Dikkes, S. J. Korsmeyer, M.E. Greenberg, Survival factor-mediated BAD phosphorylation raises the mitochondrial threshold for apoptosis, Dev. Cell 3 (2002) 631–643. [16] S.M. Dilworth, Polyoma virus middle T antigen and its role in identifying cancer-related molecules, Nat. Rev. Cancer 2 (2002) 951–956. [17] S.M. Dilworth, C.E. Brewster, M.D. Jones, L. Lanfrancone, G. Pelicci, P.G. Pelicci, Transformation by polyoma virus middle T-antigen involves the binding and tyrosine phosphorylation of Shc, Nature 367 (1994) 87–90. [18] B.J. Druker, L. Sibert, T.M. Roberts, Polyomavirus middle T-antigen NPTY mutants, J. Virol. 66 (1992) 5770–5776. [19] C.A. Dunn, V. Su, A.F. Lau, P.D. Lampe, Activation of Akt, not connexin 43 protein ubiquitination, regulates gap junction stability, J. Biol. Chem. 287 (2012) 2600–2607. [20] J. Feng, J. Park, P. Cron, D. Hess, B.A. Hemmings, Identification of a PKB/Akt hydrophobic motif Ser-473 kinase as DNA-dependent protein kinase, J. Biol. Chem. 279 (2004) 41189–41196. [21] M. Geletu, R. Arulanandam, S. Greer, A. Trotman-Grant, E. Tomai, L. Raptis, Stat3 is a positive regulator of gap junctional intercellular communication in cultured, human lung carcinoma cells, BMC Cancer 12 (2012) 605. [22] M. Geletu, C. Chaize, R. Arulanandam, A. Vultur, C. Kowolik, A. Anagnostopoulou, R. Jove, L. Raptis, Stat3 activity is required for gap junctional permeability in normal epithelial cells and fibroblasts, DNA Cell Biol. 28 (2009) 319–327. [23] M. Geletu, S. Guy, K. Firth, L. Raptis, A functional assay for gap junctional examination; electroporation of adherent cells on indium-tin oxide, J. Vis. Exp. (2014) e51710. [24] M. Geletu, S. Guy, L. Raptis, Effects of SRC and STAT3 upon gap junctional, intercellular communication in lung cancer lines, Anticancer Res. 33 (2013) 4401–4410. [25] N. Grammatikakis, A. Vultur, C.V. Ramana, A. Siganou, C.W. Schweinfest, L. Raptis, The role of Hsp90N, a new member of the Hsp90 family, in signal transduction and neoplastic transformation, J. Biol. Chem. 277 (2002) 8312–8320. [26] S. Greer, R. Honeywell, M. Geletu, R. Arulanandam, L. Raptis, Housekeeping gene products; levels may change with confluence of cultured cells, J. Immunol. Methods 355 (2010) 76–79. [27] Y.K. Hong, A. Mikami, B. Schaffhausen, T. Jun, T.M. Roberts, A new class of mutations reveals a novel function for the original phosphatidylinositol 3-kinase binding site, Proc. Natl. Acad. Sci. USA 100 (2003) 9434–9439. [28] N. Ichaso, S.M. Dilworth, Cell transformation by the middle T-antigen of polyoma virus, Oncogene 20 (2001) 7908–7916. [29] S. Ito, T. Hyodo, H. Hasegawa, H. Yuan, M. Hamaguchi, T. Senga, PI3K/Akt signaling is involved in the disruption of gap junctional communication

M. Geletu et al. / Experimental Cell Research 336 (2015) 223–231

[30]

[31]

[32]

[33]

[34]

[35] [36] [37]

[38]

[39]

[40]

[41] [42]

[43]

[44]

[45]

caused by v-Src and TNF-alpha, Biochem. Biophys. Res. Commun. 400 (2010) 230–235. S. Ito, Y. Ito, T. Senga, S. Hattori, S. Matsuo, M. Hamaguchi, v-Src requires Ras signaling for the suppression of gap junctional intercellular communication, Oncogene 25 (2006) 2420–2424. D.R. Kaplan, M. Whitman, B. Schaffhausen, L. Raptis, R.L. Garcea, D. Pallas, T. M. Roberts, L. Cantley, Phosphatidylinositol metabolism and polyoma-mediated transformation, Proc. Natl. Acad. Sci. USA 83 (1986) 3624–3628. D.R. Kaplan, M. Whitman, B.S. Schaffhausen, D.C. Pallas, M. White, L. Cantley, T. M. Roberts, Common elements in growth factor stimulation and oncogenic transformation: 85 kd phosphoprotein and phosphatidylinositol kinase activity, Cell 50 (1987) 1021–1029. R. Lin, K.D. Martyn, C.V. Guyette, A.F. Lau, B.J. Warn-Cramer, v-Src tyrosine phosphorylation of connexin43: regulation of gap junction communication and effects on cell transformation, Cell Commun. Adhes. 13 (2006) 199–216. R. Lin, B.J. Warn-Cramer, W.E. Kurata, A.F. Lau, v-Src phosphorylation of connexin 43 on Tyr247 and Tyr265 disrupts gap junctional communication, J. Cell Biol. 154 (2001) 815–827. S. Liu, S. Knapp, A.A. Ahmed, The structural basis of PI3K cancer mutations: from mechanism to therapy, Cancer Res. 74 (2014) 641–646. B.D. Manning, L.C. Cantley, AKT/PKB signaling: navigating downstream, Cell 129 (2007) 1261–1274. R. Marcellus, J.F. Whitfield, L. Raptis, Polyoma virus middle tumor antigen stimulates membrane-associated protein kinase C at lower levels than required for phosphatidylinositol kinase activation and neoplastic transformation, Oncogene 6 (1991) 1037–1040. R. Marcotte, W.J. Muller, Signal transduction in transgenic mouse models of human breast cancer—implications for human breast cancer, J. Mammary Gland Biol. Neoplasia 13 (2008) 323–335. W. Markland, S.H. Cheng, B.A. Oostra, A.E. Smith, In vitro mutagenesis of the putative membrane-binding domain of polyoma virus middle T antigen, J. Virol. 59 (1986) 82–89. S.S. Mitra, J. Xu, B.J. Nicholson, Coregulation of multiple signaling mechanisms in pp60v-Src-induced closure of Cx43 gap junction channels, J. Membr. Biol. 245 (2012) 495–506. M.S. Nielsen, A.L. Nygaard, P.L. Sorgen, V. Verma, M. Delmar, N.H. HolsteinRathlou, Gap junctions, Compr. Physiol. 2 (2012) 1981–2035. M. Pahujaa, M. Anikin, G.S. Goldberg, Phosphorylation of connexin43 induced by Src: regulation of gap junctional communication between transformed cells, Exp. Cell Res. 313 (2007) 4083–4090. L. Raptis, H.L. Brownell, K.L. Firth, L.W. MacKenzie, A novel technique for the study of intercellular, junctional communication; electroporation of adherent cells on a partly conductive slide, DNA Cell Biol. 13 (1994) 963–975. L. Raptis, H.L. Brownell, Y. Lu, T. Preston, R.P. Narsimhan, E. Schaefer, S. Anderson, T. Haliotis, v-Ras and v-Raf block differentiation of transformable C3H10T1/2-derived preadipocytes at lower levels than required for neoplastic transformation, Exp. Cell Res. 235 (1997) 188–197. L. Raptis, H.L. Brownell, A.M. Vultur, G. Ross, E. Tremblay, B.E. Elliott, Specific inhibition of Growth Factor-stimulated ERK1/2 activation in intact cells by

[46]

[47]

[48]

[49] [50] [51]

[52]

[53] [54]

[55]

[56]

[57] [58]

[59]

[60]

[61]

231

electroporation of a Grb2-SH2 binding peptide, Cell Growth Differ. 11 (2000) 293–303. L. Raptis, R. Marcellus, M.J. Corbley, A. Krook, J. Whitfield, S.K. Anderson, T. Haliotis, Cellular ras gene activity is required for full neoplastic transformation by polyomavirus, J. Virol. 65 (1991) 5203–5210. L. Raptis, A. Vultur, H.L. Brownell, E. Tomai, A. Anagnostopoulou, R. Arulanandam, J. Cao, K.L. Firth, Electroporation of adherent cells in situ for the study of signal transduction and gap junctional communication, in: S. Li (Ed.), Electroporation Protocols, The Humana Press Inc., 2008, pp. 167–183. I. Royal, L. Raptis, B.J. Druker, N. Marceau, Downregulation of cytokeratin 14 gene expression by the polyoma virus middle T antigen is dependent on c-src association but independent of full transformation in rat liver nonparenchymal epithelial cells, Cell Growth Differ. 7 (1996) 737–743. N. Smolar, B.E. Griffin, DNA sequences of polyoma virus early deletion mutants, J. Virol. 38 (1981) 958–967. J.L. Solan, P.D. Lampe, Specific Cx43 phosphorylation events regulate gap junction turnover in vivo, FEBS Lett. 588 (2014) 1423–1429. Z. Songyang, S.E. Shoelson, M. Chaudhuri, G. Gish, T. Pawson, W.G. Haser, F. King, T. Roberts, S. Ratnofsky, R.J. Lechleider, B.G. Neel, R.B. Birge, J. Eduardo Fajardo, M.M. Chou, H. Hanafusa, B.S. Schaffhausen, L.C. Cantley, SH2 domains recognize specific phosphopeptide sequences, Cell 72 (1993) 767–778. V. Su, K. Cochrane, A.F. Lau, Degradation of connexins through the proteasomal, endolysosomal and phagolysosomal pathways, J. Membr. Biol. 245 (2012) 389–400. V. Su, A.F. Lau, Ubiquitination, intracellular trafficking, and degradation of connexins, Arch. Biochem. Biophys. 524 (2012) 16–22. W. Su, W. Liu, B.S. Schaffhausen, T.M. Roberts, Association of Polyomavirus middle tumor antigen with phospholipase C-gamma 1, J. Biol. Chem. 270 (1995) 12331–12334. D.A. Talmadge, R. Freund, A.T. Young, J. Dahl, C.J. Dawe, T.L. Benjamin, Phosphorylation of middle T by pp60c-src: a switch for binding of phosphatidylinositol 3-kinase and optimal tumorigenesis, Cell 59 (1989) 55–65. C. Theiss, A. Mazur, K. Meller, H.G. Mannherz, Changes in gap junction organization and decreased coupling during induced apoptosis in lens epithelial and NIH-3T3 cells, Exp. Cell Res. 313 (2007) 38–52. E. Tomai, H.L. Brownell, T. Tufescu, K. Reid, L. Raptis, Gap junctional communication in lung carcinoma cells, Lung Cancer 23 (1999) 223–231. T. Utermark, B.S. Schaffhausen, T.M. Roberts, J.J. Zhao, The p110alpha isoform of phosphatidylinositol 3-kinase is essential for polyomavirus middle T antigen-mediated transformation, J. Virol. 81 (2007) 7069–7076. M. Whitman, D.R. Kaplan, B.S. Schaffhausen, L. Cantley, T.M. Roberts, Association of phosphatidylinositol kinase activity with polyoma mT competent for transformation, Nature 315 (1985) 239–242. M. Yoakim, W. Hou, Y. Liu, C.L. Carpenter, R. Kapeller, B.S. Schaffhausen, Interactions of polyomavirus middle T with the SH2 domains of the pp85 subunit of phosphatidylinositol-3-kinase, J. Virol. 66 (1992) 5485–5491. L. Zhou, E.M. Kasperek, B.J. Nicholson, Dissection of the molecular basis of pp60(v-src) induced gating of connexin 43 gap junction channels, J. Cell Biol. 144 (1999) 1033–1045.