Skip Navigation


Carcinogenesis Advance Access originally published online on June 15, 2006
Carcinogenesis 2006 27(12):2528-2537; doi:10.1093/carcin/bgl110
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
27/12/2528    most recent
bgl110v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (6)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Kalra, J.
Right arrow Articles by Laird, D. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kalra, J.
Right arrow Articles by Laird, D. W.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2006. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Cx26 inhibits breast MDA-MB-435 cell tumorigenic properties by a gap junctional intercellular communication-independent mechanism

Jessica Kalra1, Qing Shao, Hong Qin2, Tamsin Thomas, Moulay A. Alaoui-Jamali3 and Dale W. Laird*

Department of Anatomy and Cell Biology, University of Western Ontario London, Ontario, Canada N6A 5C1
1 British Columbia Cancer Research Center, Vancouver BC, Canada V5C-1L3
2 Department of Lymphoma and Myeloma, MD Anderson Cancer Center Houston TX 77030, USA
3 Department of Medicine, Pharmacology and Therapeutics and Centre for Translational Research in Cancer, Lady Davis Institute for Medical Research, McGill University Montreal, Quebec, Canada H3T 1E2

*To whom correspondence should be addressed. Tel: +1 519 661 2111, Ext: 86827; Fax: +1 519 850 2562; Email: dale.laird{at}schulich.uwo.ca


   Abstract
Top
 Abstract
Introduction
 Materials and methods
 Results
 Discussion
 References
 
It has been well established that the restoration of connexin expression in tumor cells often leads to a partial reversion of tumor cell phenotypes and increased growth control. In this study, a less-aggressive variant of MDA-MB-435 cells obtained from the MDA-MB-435 cell line was engineered to express gap junctional intercellular communication (GJIC)-competent Cx26, a GJIC-incompetent cell surface transported GFP-Cx26 chimera, or a Golgi apparatus-localized, disease-linked Cx26 mutant (D66H). Collectively, these cell lines were designed to establish whether Cx26 regulates tumor properties, such as migration, invasion and growth, by (i) a GJIC-dependent pathway; (ii) a mechanism requiring Cx26 transport to the cell surface; or (iii) a mechanism where Cx26 expression alone was sufficient. The expression of Cx26 and green fluorescent protein (GFP)-Cx26 decreased cell proliferation while all three Cx26 variants inhibited anchorage-independent cell growth. All three Cx26 variants also altered the distribution of filamentous actin and significantly reduced cell migration, while only the D66H mutant failed to inhibit cell invasion through matrigel. Furthermore, expression of all the Cx26 variants reduced the levels of total ß1 integrin, and decreased the activity of matrix metalloproteinase-9 (MMP-9) while increasing tissue inhibitors of MMP-1 (TIMP-1) activity. Interestingly, the expression of Cx43 regulated the same gene products without significantly affecting the tumorigenic properties of the MDA-MB-435 cells. Together, these results suggest that Cx26 expression, independent of the necessity for gap junctional intercellular communication, partially reverted MDA-MB-435 cell properties associated with tumorigenesis, and regulated the expression of genes important in cell migration and invasion.

Abbreviations: Cx, connexin; D66H-GFP, deafness and skin disease-linked Cx26 mutant; GFP, green fluorescent protein; GJIC, gap junctional intercellular communication; MMP, matrix metalloproteinase; TIMP, tissue inhibitors of MMPs


   Introduction
Top
 Abstract
 Introduction
Materials and methods
 Results
 Discussion
 References
 
Gap junctions are clusters of membrane spanning channels that link neighboring cells and allow for the direct intercellular exchange of ions, small metabolites and second messengers (1). A single gap junction channel consists of two hemi-channels called connexons, and each connexon is composed of six transmembrane proteins known as connexins (2). Connexins are a family of proteins composed of 21 members that are differentially expressed in almost all tissues (3). Gap junctional intercellular communication (GJIC) provides a mechanism for individual cells to communicate directly with each other and regulate important cellular functions ranging from homeostasis to highly specialized cell activities (2). Important to the current study are the numerous reports that have linked the loss of GJIC and/or connexin expression to tumor cell growth (see ref. 4 for a review). Cx26 and Cx43 re-introduction into many rodent and human breast tumor cell lines has reverted the tumor phenotype, exemplified by the establishment of significant levels of growth control both in vitro and in vivo (5–17). Pertinent to this study, both Cx26 and Cx43 restored the differentiation potential and growth control of MDA-MB-435 cells (16).

In earlier studies, it was assumed that connexins would exert their tumor growth suppression effect through their primary role in establishing GJIC and the direct exchange of secondary messengers and other small molecules. However, the number of possible mechanism(s) responsible for connexin-mediated tumor suppression has grown with recent insights into the possible functional roles of hemichannels (18), the cross-talk between gap junctions and other adhesion-based junctional complexes (19–22) and the discovery of novel connexin-binding proteins (23–25). Thus, there are several putative mechanisms by which connexins may participate in regulating cell growth and differentiation that are GJIC-independent (26–29). Furthermore, connexins have been shown to regulate a variety of molecules involved in growth and proliferation (30), tumor angiogenesis (8) and cell cycle control (31).

Although many studies have examined the role of connexins in regulating cell growth within the primary tumor, far less is known about the role of connexins and gap junctions during the various stages of metastasis. Reports have suggested that connexin expression reduces the likelihood and instances of tumor cell metastasis (32–34). Weakly metastatic rat mammary clones were shown to have higher incidences of GJIC and more gap junction plaques than highly metastatic clones (35). In addition, quite interesting is the evidence that under the influence of metastasis suppressive proteins such as breast metastasis suppressor-1 (BRMS-1) (34) or tissue inhibitors of MMPs-1 (TIMP-1) (36), there was decreased metastasis as well as a restoration of connexin content and/or GJIC. Relevant to our study, the expression of BRMS-1 was shown to cause an increase in Cx43 expression in MDA-MB-435 cells (34). However, other reports have documented that gap junctions form between tumor cells and endothelium (32,37) and one report has suggested that the expression of Cx26 may facilitate melanoma cell metastasis in vivo (38). Thus, a relationship between connexin expression and metastatic potential is beginning to emerge but remains controversial owing to the complexity of the multiple stages and environments tumor cells must pass before colonizing in a distinct organ. In addition, it is also be coming more clear that different members of the connexin family as well as different tumor cell types may engage different mechanisms to either inhibit or assist the molecular processes needed for cells to metastasize.

In the current study, we used a variant of the human breast tumor cell line, MDA-MB-435, to examine if the expression of functional and GJIC-incompetent species of Cx26 could change cell growth characteristics, migration and invasion potential of the cells and regulate genes important in cancer progression. To this end, MDA-MB-435 cells were engineered to express (i) functional wild-type Cx26; (ii) a GJIC-incompetent variant of Cx26 that is transported to the cell surface; and (iii) a Cx26 mutant found in human populations with Vohwinkels syndrome, which resides within the Golgi apparatus and is therefore also non-functional in terms of GJIC.


   Materials and methods
Top
 Abstract
 Introduction
 Materials and methods
Results
 Discussion
 References
 
Cell culture
MDA-MB-435 and NRK (normal rat kidney) cells were cultured in Dulbecco's modified Eagle's high glucose medium obtained from Invitrogen (Burlington, ON) and MDA-MB-231 cells were cultured in RPMI 1640 medium from Sigma-Aldrich (St Louis, MO). All cell lines were originally obtained from the American Type Culture Collection (Monassas, VA). Cells were maintained at 37°C in an environment of 95% air and 5% CO2 in the recommended medium supplemented with 10% fetal bovine serum (FBS), 100 U/ml penicillin, 100 µg/ml streptomycin and 2 mM glutamine.

Constructs and retroviral infection
Generation of AP2 retroviral vectors containing cDNA for Cx43, Cx26, green fluorescent protein (GFP)-Cx26 or D66H-GFP were described previously (7,39,40). All four recombinant retroviral vectors, as well as an empty retroviral vector, were transfected into the 293GPG packaging cell line to produce replication-defective virus-containing supernatant, and the retroviral infection of MDA-MB-435 cells was accomplished as described previously (41). Five cell lines were created: MDA-MB-435 cells expressing a functional Cx26 (Cx26), a GJIC-incompetent GFP-tagged Cx26 (GFP-Cx26), a mutant Cx26 (D66H-GFP), a functional Cx43 (Cx43) and an empty viral vector (v) control. In each case, the expression of recombinant connexins exceeded 90% of the cultured cells as revealed by immunofluorescent labeling for connexins or intrinsic GFP fluorescence. The functional status of MDA-MB-435 cells that expressed wild-type Cx26 and GFP-tagged Cx26 was assessed in a previous study through the use of gap junction permeable dyes (8).

Immunofluorescent labeling and confocal microscopy
Cells were immunolabeled as described previously by Laird et al. (42). Antibodies used in this study included a polyclonal anti-Cx43 (Sigma-Aldrich) (1:500), a polyclonal anti-Cx26 (Zymed Laboratories Inc., San Francisco, CA) (1:200), or a polyclonal anti-giantin (BabCo, Richmond, CA) (1:1000). Secondary antibodies conjugated to either Texas Red (1:200) or FITC (1:100) (Jackson ImmunoResearch Lab, West Grove, PA) were used to detect antibody binding. Phalloidin conjugated to Texas Red was used to label filamentous actin at a dilution of 1:50 (Molecular Probes, Eugene, OR). In many instances cells were labeled with Hoechst 33342 (final concentration 10 µg/ml; Molecular Probes) for 15 min.

Immunofluorescently labeled cells were imaged on a Zeiss (Thornwood, NY) LSM 510 META confocal microscope mounted on an inverted Axiovert 200 motorized microscope equipped with a x63 oil (1.4 numerical aperture) objective. FITC was excited using a 488 nm argon laser and resulting fluorescence was collected through a 500–550 nm band pass filter. A 543 nm helium–neon laser was used to excite Texas Red and emission collected after passage through a 590 nm long pass filter. To acquire images of the Hoechst 33342 nuclear stain, a Chameleon multi-photon laser set at 730 nm was used and the resulting fluorescence was acquired after passage through a 390–465 nm band pass filter.

Cell growth in soft agar
A 12-well plate was coated with 0.4% soft agar (ICN Biomedical, Aurora, OH) diluted in cell culture media and allowed to solidify at room temperature. Cells (3.75 x 104 cells/ml) were added to 0.3% soft agar diluted in Dulbecco's modified Eagle's high glucose medium. This mixture was seeded on top of the agar-coated wells and allowed to solidify. An extra 300 µl of culture media was applied to the well and changed every other day for 10 days. The colonies (>10 cells) were counted in 10 random fields with a standard inverted Leica DM IL microscope (Leica Microsystems Canada, Richmond Hill, ON) using a x10 objective lens. The investigator was blinded to the cell treatments being counted. The experiment was repeated four times, with three internal replicates. Data are expressed as mean ± SD and parametric analysis was performed using an unpaired Student's t-test. All statistical data were calculated using GraphPad Prism (San Diego, CA).

Cell growth analysis
Cells (2.5 x 104) were suspended in 1.0 ml of culture medium and plated in a 6- well dish with an additional 1 ml of culture media. Cells were collected and counted at Days 3, 5, 7 and 9 after plating. The experiment was repeated three times, with three internal replicates. Data are expressed as mean ± SD and parametric analysis was performed using an unpaired Student's t-test.

Trans-well migration and invasion assays
Cells were prepared in serum-free cell culture media to a final concentration of 1 x 105 cells/ml. Culture media containing 10% fetal bovine serum (FBS), as chemo-attractant, was added to all the wells of the 24-well plate. A FluorBlok insert with a porous membrane (0.8 µm pore size) (BD Bioscience Discovery Labware, Canaan, CT) was transferred to each well containing the chemo-attractant, creating an upper and lower chamber. Cells (5 x 104) were placed into the upper chamber and incubated for 24 h at 37°C. Invasion studies were completed as stated above over a 48 h period using FluorBlok membranes pre-coated with 62.5 µg/ml of Matrigel (Bioscience, Canaan, CT) diluted in cell culture media. Insert membranes were pre-coated and allowed to set for 1 h in a tissue culture incubator before the addition of the cells and media.

To assess the extent of cell migration or invasion, cells adhering to both the top and the bottom of the membrane were rinsed with Hank's balanced salt solution (HBSS) (Invitrogen, Burlington, ON). Cells were fixed with HemaColor solution 1 fixative (EM Science Harleco, Kansas City, MO). Nuclei on both the top and bottom of each membrane were stained with 10 µg/ml Hoechst 33342. Random digital images (10 from the top and 10 from the bottom of the filter) were acquired using Leica microscope equipped with a digital CCD camera (Hamamatsu Photonics, Japan). Acquired images were density sliced and Open Lab software (Improvision, Lexington, MA) was programmed to calculate the total area occupied by stained nuclei per image. This area was divided by the average area of a single nucleus (~0.4 µm2), to estimate the number of cells in each image. The experiment was repeated four times, with three internal replicates. Data are expressed as mean ± SD and parametric analysis was performed using an unpaired Student's t-test.

SDS–PAGE and western blot
SDS–PAGE and western blotting was performed as described previously (42). The blots were labeled with anti-Cx43 (Sigma-Aldrich) (1:5000 dilution), rabbit polyclonal anti-Cx26 [Zymed Laboratories, Z-Z8 (1:200) or UM214 (1:1000)], or anti-ß1 integrin (Santa Cruz Biotechnology, Santa Cruz, CA) (1:200) antibodies. Primary antibody binding was detected by further incubations with anti-mouse, or anti-rabbit horseradish peroxidase (HRP) (1:10 000) (Pierce Biotechnology, Rocklord, IL) and chemiluminescence (Pierce Biotechnology) To assess equal protein loading, nitrocellulose membranes were stripped in REblot (Chemicon International, Temecula, CA), or replica gels were probed for GAPDH (1:10 000) (Cedarlane Laboratories, Hornby, ON, Canada). Immunoblots were scanned and quantified using Kodak Digital Science 2.0 (Eastman Kodak, Rochester, NY). Data are expressed as mean ± SD and parametric analysis was performed using an unpaired Student's t-test.

Zymography and reverse zymography
Equal numbers of cells were grown to 80% confluence, and then incubated with 2 ml of serum free culture media. After 48 h, media were collected, centrifuged at 10 000 g for 10 min at 4°C, and the supernatant was assessed for protein concentration using the BCA protein assay. Protein (40–50 µg) was diluted in non-reducing sample buffer [0.5 M Tris (pH 6.8), 20% glycerol, 4% SDS and 0.005% bromophenol blue]. Zymography was carried out as described previously with the following changes (43,44). Equal concentrations of proteins were separated by 10% SDS–PAGE containing 1 mg/ml gelatin substrate (ICN Biomedical, Aurora, OH). Upon completion of electrophoresis, gels were washed twice in 2.5% Triton X-100 solution and rinsed in distilled, deionized H2O. The gels were incubated for up to 36 h at 37°C in a developing buffer [50 mM Tris (pH 7.4), 0.2 M NaCl, 5 mM CaCl2 and 0.02% Brij 35]. Gels were stained in Coomassie Brilliant Blue (Bio-Rad Laboratories, Hercules, CA) and zymographs were analyzed and quantified for densitometry using Kodak Digital Science 2.0 (Eastman Kodak, Rochester, NY). The experiment was repeated four times. Data are expressed as mean ± SD and parametric analysis was performed using an unpaired Student's t-test. Reverse zymography was preformed as described above with a 12% SDS–PAGE where the gel included 1 ml of purified MMP-9 and included a control lane of purified TIMP-1 (Dr Kevin Leco, UWO, London, ON, Canada).


   Results
Top
 Abstract
 Introduction
 Materials and methods
 Results
Discussion
 References
 
Expression of connexin variants in MDA-MB-435 cells
In order to assess the mechanistic role of connexins in the regulation of cell migration, growth and invasion, it was necessary to generate a series of connexin-positive cell lines on a genetically identical background. To that end, we initiated these studies by characterizing the growth and phenotypic properties of an MDA-MB-435 human breast tumor cell line. Surprisingly, these cells were found to be far less aggressive than reported in other studies (16) exhibiting minimal ability to form tumors in nude mice (data not shown) and they exhibited reduced invasive properties when grown on or within matrigel-coated filters as compared with other aggressive human breast tumor cell lines such as MDA-MB-231 (Figure 4B). In addition, although MDA-MB-435 cells have been reported to express both Cx43 and Cx32 (34) we found no evidence for Cx43 protein and only a small subpopulation of cells expressed any detectable Cx32 protein. Nevertheless, this less aggressive ‘variant’ of the MDA-MB-435 cell line was devoid of Cx26, GJIC-deficient and sufficiently migratory and invasive to assess the impact of the overexpression of functional and GJIC-incompetent connexin species. Importantly, these cells were efficient at trafficking overexpressed connexins to the cell surface for the formation of functional gap junctions, properties not exhibited by MDA-MB-231 cells (7).

Immunolabeling (Figure 1A) and western blots (Figure 1B) revealed that GJIC-incompetent MDA-MB-435 (WT) and cells infected with an empty viral vector (v) are devoid of detectable Cx26 (Figure 1A) and Cx43 (Figure 1B). Cells expressing wild-type Cx26 or Cx43 (Figure 1A, arrows) formed gap junction-like structures at the cell surface and appropriately sized protein bands were detected by western blot (Figure 1B). It is notable that multiple phosphorylation species of Cx43 were identified by western blot at 41–44 kDa but since Cx26 is not phosphorylated, only one protein band was detected as expected at ~21 kDa. When functionally inactive GFP-tagged Cx26 was expressed in MDA-MB-435 cells, apparent gap junction plaques were observed at the cell surface (Figure 1A, arrows) and immunoblotting revealed an appropriately sized fusion protein band at ~48 kDa (Figure 1B). Conjugation of the GFP tag to the N-terminal of Cx26 resulted in the formation of non-functional plaques in MDA-MB-435 cells (8). The deafness and skin disease-linked Cx26 mutant, D66H-GFP, was localized to the Golgi apparatus as revealed by the co-localization between D66H-GFP and resident Golgi protein, giantin (Figure 1A, insert, note yellow color). Since the Cx26 antibody used in this study did not bind human Cx26, the D66H-GFP mutant was not detected on the western blot (Figure 1B). Consequently, cell lysates were probed with a second anti-Cx26 antibody to reveal the presence of the GFP-tagged Cx26 and D66H mutant (Figure 1C). This study revealed that, although the D66H mutant was expressed in >90% of the cell population, it was far less abundant than GFP-Cx26 assuming the antibody had equal affinity for both rat and human Cx26. As controls, Cx43 was localized within gap junction plaques in NRK cells (Figure 1A) but predominantly found in intracellular structures in MDA-MB-231 cells as described previously (Figure 1A) (8).


Figure 1
View larger version (61K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 Expression patterns of untagged and GFP-tagged connexins in mammalian breast tumor cell lines. MDA-MB-435 wild-type (WT) cells, cells infected with an empty retroviral vector (v) or MDA-MB-435 cells expressing Cx26 (Cx26) or Cx43 (Cx43) were immunolabeled and immunoblotted for Cx26 or Cx43 (A and B). Note that wild-type or empty vector control cells lacked Cx26 but cell surface gap junction plaques were observed in cells overexpressing Cx26 or Cx43 (A, arrows). Similarly, gap junction plaques were readily observed in MDA-MB-435 cells expressing GFP-tagged Cx26 (GFP-Cx26) (A, arrows) whereas a disease-linked Cx26 D66H mutant (D66H-GFP) was retained within a perinuclear compartment that co-localized with the resident Golgi protein, giantin (A, insert, yellow). Cx43 was retained within an intracellular compartment in MDA-MB-231 cells (A) but assembled into gap junctions at cell–cell interfaces in normal rat kidney (NRK) cells (A). All cells, except NRK, were counterstained with Hoechst 33342 to denote the nuclei. Bar = 10 µm. Cx26 and GFP-Cx26 were identified by western blots in appropriate cell types whereas GAPDH was used as a loading control (B). A second anti-Cx26 antibody that recognizes both human and rat Cx26 was used to probe cell lysates obtained from cells overexpressing GFP-Cx26 or D66H-GFP (C).

 
Expression of functional and GJIC-incompetent Cx26 decreases the tumorigenic characteristics of MDA-MB-435 cells
In vitro cell growth analysis revealed that cells expressing Cx26 or GFP-Cx26 experienced a lag in proliferation and by Day 9, a significant suppression of growth rate was observed as both cell lines had ~40% less cells than the vector control (Figure 2A). Surprisingly, neither Cx43 nor D66H-GFP expressing cells inhibited growth in vitro (Figure 2B), suggesting that the suppression of tumor cell growth was restricted to Cx26 when expressed at the cell surface and assembled into apparent gap junction plaques. Interestingly, upon the expression of Cx26, GFP-Cx26 and D66H-GFP, MDA-MB-435 cells had a 76, 68 and 38% reduced frequency of colony formation in soft agar, respectively, when compared with the vector control cells. The expression of Cx26, and GFP-Cx26, also had an effect on colony size, rarely forming colonies with ≥10 cells.


Figure 2
View larger version (17K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 Cx26-induced inhibition of MDA-MB-435 cell growth and growth in soft agar. Growth suppression was observed in cells expressing Cx26 (closed triangles) or GFP-Cx26 (open triangles) but not in wild-type cells (closed circles) cells containing the empty vector (open circles) or cells expressing D66H-GFP (closed squares) or Cx43 (open squares) (A). Cells expressing Cx26, GFP-Cx26 or D66H-GFP exhibited a reduction of cell growth and colony formation in soft agar compared with empty virus-infected cells (B). Growth curves were expressed as mean ± SE (n = 4) (*P < 0.05, ***P < 0.001). Bar = 10 µm.

 
MDA-MB-435 cells expressing connexins exhibited altered morphology but only cells expressing Cx26 variants were less migratory and invasive
Phalloidin-staining of actin microfilaments revealed that wild-type and cells infected with an empty vector exhibited a flattened phenotype, with dispersed and periphery-localized stress fibers (Figure 3). However, upon the expression of Cx26, GFP-Cx26, D66H-GFP or Cx43, MDA-MB-435 cells exhibited a more compact shape, with many overlapping processes and filamentous actin staining was primarily perinuclear (Figure 3, arrows). Interestingly, Cx26, GFP-Cx26 or D66H-GFP overexpression significantly inhibited MDA-MB-435 migration compared with wild-type or empty vector control cells (P < 0.01) (Figure 4A). The percentage of migrating cells seen in all three Cx26-expressing cell lines was similar to that seen in both normal mammary epithelial cells, MCF10A, which revealed minimal migration after a 24 h period (data not shown), and non-transformed, non-motile normal rat kidney (NRK) cells. Notably, there was no significant reduction in Cx43 expressing MDA-MB-435 cell migration. To examine whether Cx26 expression reduced the ability of MDA-MB-435 cells to digest and pass through a thin layer of matrigel, invasion assays were performed in vitro. First, it is notable that MDA-MB-435 cells were about one-third as efficient in invading though the matrigel coating as compared with MDA-MB-231 cells. Our results revealed that MDA-MB-435 cells expressing Cx26 or GFP-Cx26 were significantly less invasive than the controls (P < 0.05) (Figure 4B). Although D66H-GFP cells also trended toward a reduced capacity for invasion, this inhibition was not statistically significant (Figure 4B).


Figure 3
View larger version (91K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 Expression of connexin variants induced a perinuclear distribution of F-actin as revealed by phalloidin staining. Wild-type (WT) MDA-MB-435 cells, cells infected with an empty retroviral vector (v) or cells expressing Cx26, Cx43, GFP-Cx26 or D66H-GFP were stained with phalloidin conjugated to Texas Red. The perinuclear arrangement of F-actin revealed a spindle shaped cytoskeletal structure in cells expressing Cx26, GFP-Cx26, D66H-GFP or Cx43. WT and v cells had a flattened appearance with actin stress fibers in the cortical regions and in association with the cell surface. Bar = 10 µm.

 

Figure 4
View larger version (14K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4 Overexpression of Cx26 variants decreased cell migration and invasion. Empty virus control cells (v) show no statistically significant difference in migration or invasion from WT cells, whereas cells expressing Cx26, GFP-Cx26 or D66H-GFP were less migratory compared with empty vector control cells (P < 0.01) (A). Cx26 and GFP-Cx26 expressing cells were less invasive compared with vector control cells (P < 0.05) (B). D66H-GFP also showed a considerable reduction in cell invasion although this change was not statistically significant (n = 4) (*P < 0.05, **P < 0.005).

 
ß1 Integrin expression, TIMP-1 and MMP-9 activity were regulated in connexin-expressing cells
Our previous Affymatrix array studies suggested that ß1 integrin was downregulated in cells expressing Cx26 or GFP-Cx26 compared with control cells whereas TIMP-1 expression was increased (Qin et al., unpublished data). In accordance with this array data, western blots revealed that ß1 integrin expression was ~3-fold lower in cells engineered to express functional, GJIC-incompetent or mutant Cx26 as well as Cx43 (Figure 5A and B).


Figure 5
View larger version (18K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5 ß1-Integrin was downregulated in MDA-MB-435 cells expressing wild-type connexins and GJIC-incompetent Cx26 variants. Western blots revealed that ß1-integrin was downregulated in cells expressing Cx43, Cx26, GFP-Cx26 or D66H-GFP in comparison with wild type (WT) or empty viral vector controls (v) (A). Quantification of three independent experiments revealed a significant reduction in ß1-integrin expression in cells engineered to express any of the wild-type or connexin variants. No statistical difference was seen between WT and v cells (B) (n = 3) (**P < 0.005, ***P < 0.001).

 
Zymography and reverse zymography were carried out in order to determine the activity level of both the inhibitor TIMP-1, and the enzyme, MMP-9, in MDA-MB-435 cells expressing Cx26, GFP-Cx26 or D66H-GFP as compared with control cells. Gelatin zymography revealed a significant reduction in MMP-9 activity in cells expressing Cx26, GFP-Cx26, D66H-GFP or Cx43 (Figure 6A and C). Coordinately, reverse zymography revealed an increase in TIMP-1 activity in cells expressing Cx26, GFP-Cx26, D66H-GFP or Cx43 as seen by increased substrate digestion by these cell lines (Figure 6B). These results suggest that both TIMP-1 and MMP-9 are regulated by all functional and GJIC-incompetent connexin variants used.


Figure 6
View larger version (35K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 6 Decreased MMP-9 activity and a corresponding increase in TIMP-1 activity were observed in cells expressing wild-type connexins and Cx26 variants. Gelatin zymography of cultured media collected from wild-type (WT) cells and cells expressing the empty vector (v), or cells expressing Cx26, GFP-Cx26, D66H-GFP or Cx43 was used to assess changes in MMP-9 activity (A). Although, proMMP-9 (92 kDa) and active MMP-9 (84 kDa) were detected in all cells, a decrease in MMP-9 activity was seen in cells expressing Cx26, GFP-Cx26, D66H-GFP or Cx43 (C) (n = 4). Reverse gelatin zymography of the same cultured media was completed in order to determine if there was an increase in TIMP-1 activity in connexin-expressing cells. Increased TIMP-1 (migrating at 28 kDa) activity (note the reduction in band intensity within the red box area) was seen upon the expression of functional, GJIC-incompetent and mutant Cx26 as well as Cx43 (B) (n = 3) (*P < 0.05, **P < 0.005).

 

   Discussion
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
References
 
Connexins as tumor suppressors
The purpose of this study was to elucidate the mechanism(s) by which connexins act as tumor suppressors. In principle, there are four basic mechanisms by which connexins may suppress tumor growth: (i) GJIC, (ii) functional hemichannels, (iii) gap junction enhanced intercellular adhesion and (iv) connexin-dependent modulation of proteins that regulate tumor cell behavior. It is also plausible that two or more of these mechanisms may act in combination depending on the type of tumor resulting in both cell-type and connexin-specific mechanisms of tumor suppression. It is notable that two of these four mechanisms would not necessitate that connexins form functional gap junction channels or hemichannels. This study focuses on Cx26 because it is expressed in the luminal cells of human as well as rodent breast lobules and ducts (5,45) which are the origin of most breast tissue carcinomas. Importantly, Cx26 was identified as a tumor suppressor in both rodent and human breast tumors and cell lines (7,13–16). Thus, the purpose of this study was to engineer and employ a series of cell lines which would allow for the investigation of the mechanism by which Cx26 may act to suppress the tumorigenic phenotypes of human breast cancer cells. A less aggressive variant of the connexin-deficient human breast tumor cell line, MDA-MB-435, was chosen as a platform to re-introduce functional and GJIC-incompetent variants of Cx26.

GJIC appears unnecessary for Cx26-based growth control
It has already been established that connexins exert an inhibitory effect on tumor growth when re-introduced in breast tumor cells; however, the properties of Cx26 that are required to exert this effect have remained elusive. Our studies revealed that the cell surface expression of a GJIC-incompetent variant of Cx26 (GFP-Cx26) possessed a similar ability as functional Cx26 to inhibit in vitro growth of MDA-MB-435 cells in the absence of any contribution to GJIC or functional hemichannel formation (8,46,47). With increasing evidence of cross-talk between gap and adherens junctions (25), one possibility is that cell surface Cx26 contributes to the overall adhesive properties between adjacent cells. Cell adhesion is thought to be required for the assembly of gap junctions at the cell surface (48). Conversely, blocking gap junction assembly has been shown to inhibit the formation of adherens junctions in rat hepatoma cells (19). It has also been suggested that connexins may interact with other junctional complexes to regulate overall cell adhesion and intercellular junction formation (49). A GJIC-independent but connexin-dependent pathway for regulating cell growth has gained increasing merit as several studies support this possibility (7,26–29). Since the mere expression of Cx26 alone in the form of the D66H-GFP mutant did not reduce MDA-MB-435 cell growth, this would suggest that the cell surface expression of Cx26 may be a critical necessity. Although such a conclusion would seem reasonable some caution should be exercised before reaching such a position. For instance, even though the D66H mutant was expressed in the vast majority of cells the overall expression level was significantly less compared with either Cx26 or GFP-Cx26, thus leaving the possibility that at high levels the D66H mutant may also reduce cell growth. Moreover, our studies revealed that Cx43 can readily assemble gap junction plaques but failed to reduce MDA-MB-435 cell growth further suggesting that the mere delivery of ‘any’ connexin is not sufficient to induce growth control.

Cx26 inhibits migration and invasion independent of its cell surface transport and assembly into functional gap junctions
In our study, we present evidence that the functional status or subcellular localization of Cx26 appears to be of minimal significance for Cx26 to inhibit MDA-MB-435 cell migration and invasion or to regulate genes important in these processes. In this regard, all overexpressed Cx26 variants exhibited the ability to significantly inhibit the migration of MDA-MB-435 cells. At this stage it is impossible to determine if they regulate migration by identical mechanisms and given that the D66H mutant has a steady-state profile that reflects its localization to the Golgi apparatus (40), it is possible that the bolus of mutant protein may act to indirectly affect the secretion of proteins important in migration. Given the immense capability of cells to secrete, this seems unlikely but all protein overexpression strategies remain subject to such critique. It is also difficult to completely eliminate the possibility that Cx26 variant overexpression may upregulate other compensatory connexins family members in these MDA-MB-435 cells. However, we have shown previously that Lucifer yellow failed to transfer between MDA-MB-435 cells that overexpressed GFP-Cx26 (8). It is notable that the D66H mutant caused a non-statistically significant trend toward a reduction in cell invasion. At first approximation, this could suggest that it is a far less potent mutant (or even incompetent) at inhibiting cell invasion but this may also reflect the fact that it is expressed at lower levels than either Cx26 or GFP-Cx26. It is also important to note that we examined the migratory or invasive properties of cells that have disrupted their junctional contact with neighboring cells before movement through the pores of uncoated or matrix-coated filters. Given the well understood principle that tumor cell lines are typically heterogeneous, it is possible that cells that retain junctional contact with their neighbors, represent a second population that are more resistant to Cx26-induced changes in cell migratory and invasive properties.

In culture, MDA-MB-435 cells exhibited characteristics of integrin-induced, actin-dependent cytoskeletal remodeling (50,51) and cell spreading as evidenced by prominent stress fibers. Cells expressing functional, GJIC-incompetent and mutant Cx26 appeared more compact with actin staining being mainly perinuclear. In prostate cells, drug inhibition of Cx26-based GJIC was found to inhibit invasion and migratory properties of PC-3 cells, and furthermore, evidence was presented for a direct interaction between Cx26 and focal adhesion kinase (52). This study would argue for a link between Cx26 and focal contacts through a mechanism that may not require Cx26 to be functional in cell–cell communication. Moreover, other studies have shown that interactions with the extracellular matrix induce GJIC and promote mammary epithelial cell differentiation (53). Interestingly, a recent study demonstrated a potential role for ß1-integrin in the development of breast cancer in conditional transgenic mouse model (54). In our study, western blot analysis revealed that the expression of functional, GJIC-incompetent and mutant Cx26, as well as Cx43, downregulated ß1-integrin levels by 2- to 3-fold. Hence, the Cx26-induced cytoskeletal rearrangement and reduction in ß1-integrin expression seen in MDA-MB-435 cells may be part of the molecular mechanism responsible for the phenotypic change in cell migration. However, since Cx43 expression also induced a downregulation of ß1-integrin with only a minimal effect on cell migration and invasion, the mechanism must be more complex and extend beyond ß1-integrin regulation.

MMP-9, an enzyme involved in extracellular matrix degradation (55) and a direct substrate of TIMP-1 (56), was found to have lower activity levels in all Cx26 and Cx26 variant expressing cells when compared with controls. Changes in MMP-9 activity levels in combination with the levels of TIMP-1, which bind and inactivate MMPs, would be expected to affect the ability of MDA-MB-435 tumor cells to invade surrounding extracellular matrix. In breast cancer cells there are increased levels of MMP-1, -2 and -9 in both the cancerous and surrounding non-cancerous cells (57). The activation and expression of MMP-9 has been shown to be related to a number of different molecules, that include ß1-integrin (58) and TIMP-1 (56), both of which have been demonstrated in this study, to be regulated by Cx26 and Cx43 expression in MDA-MB-435 cells.

Only Cx26 variants regulated MDA-MB-435 tumorigenic cell properties yet common genes were regulated by both Cx26 and Cx43!
The lack of a significant Cx43-linked inhibition of growth, migration and invasion in MDA-MB-435 revealed that not all connexins exhibit the same properties in vitro. The obvious conclusion form these studies is that ß1-integrin, MMP-9 and TIMP-1 represent an incomplete set of genes regulated by Cx26 and other genes must be coordinately altered by Cx26 that are not regulated by Cx43. In many ways this is not unexpected given the multitude of differences between connexin family members related to binding proteins (24) and channel permeability characteristics (1). Previous studies using GJIC-deficient cervical HeLa cells showed that although GJIC was restored upon the overexpression of a variety of connexins including Cx40, Cx43 and Cx26, only Cx26 expressing cells demonstrated decreased growth rates in vitro and reduced tumor formation in vivo (59). It is also notable that some connexins are poor tumor suppressors in vitro but potent inhibitors of tumor cell growth in vivo (7).

How does GJIC-incompetent Cx26 signal to regulate gene expression levels?
Mechanistically, it would seem plausible that Cx26 and its GJIC-incompetent variants regulate cell proliferation, migration and invasion by putative binding partners. It has already been established that more than one member of the connexin family of proteins interacts with ZO-1, ZO-2, Src, tubulin and several protein kinases (24,25). However, only a ubiquitin ligase subunit, OCP1, found in the organ of Corti (60) and focal adhesion kinase (52) have been reported to interact with Cx26. Cx26 has a relatively short polypeptide domains exposed to the cytoplasm and lacks any nuclear localization sequences. Thus, it is most unlikely that any Cx26 proteolytic fragment or intact molecules extracted from the membrane would signal directly to the nucleus. Given the surge of new connexin-binding partners being identified it is likely that additional Cx26-associated signaling molecules will be identified in the near future that may shed some insight into how Cx26 is linked to tumor growth, progression and metastasis.

In summary, these data indicate that Cx26 plays a pivotal role in reducing the tumorigenicity of a non-aggressive variant of the MDA-MB-435 breast tumor cell line. Cx26-induced tumor suppression relies on a unique GJIC-independent mechanism, although it still remains possible that GJIC will augment the tumor suppressive effect or that GJIC will play a more central role in other cell types. Interestingly, the overexpression of Cx26 variants, whether intracellular localized or at the cell surface, reduced the migration and invasion of MDA-MB-435 cells. A unifying feature of the regulation evoked by these Cx26 variants was their consistent regulation of ß1-integrin, MMP-9 and TIMP-1 potentially providing insight into how Cx26 may regulate both MDA-MB-435 cell migration and invasion. The observed molecular changes may represent only a few of a vast number of molecules that are regulated by the expression of Cx26 in breast tumor cells.


   Acknowledgments
 
This study was supported by the Canadian Breast Cancer Research Alliance grants (09233) to D.W.L. and M.A.J.

Conflict of Interest Statement: None declared.


   References
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 

  1. Goldberg G.S., Valiunas V., Brink P.R. (2004) Selective permeability of gap junction channels. Biochim. Biophys. Acta 1662:96–101.[Medline]
  2. Saez J.C., Berthoud V.M., Branes M.C., Martinez A.D., Beyer E.C. (2003) Plasma membrane channels formed by connexins: their regulation and functions. Physiol. Rev. 83:1359–1400.[Abstract/Free Full Text]
  3. Sohl G. and Willecke K. (2004) Gap junctions and the connexin protein family. Cardiovasc. Res. 62:228–232.[CrossRef][ISI][Medline]
  4. Naus C.C. (2002) Gap junctions and tumour progression. Can. J. Physiol. Pharmacol. 80:136–141.[CrossRef][ISI][Medline]
  5. Jamieson S., Going J.J., D'Arcy R., George W.D. (1998) Expression of gap junction proteins connexin 26 and connexin 43 in normal human breast and in breast tumours. J. Pathol. 184:37–43.[CrossRef][ISI][Medline]
  6. Laird D.W., Fistouris P., Batist G., Alpert L., Huynh H.T., Carystinos G.D., Alaoui-Jamali M.A. (1999) Deficiency of connexin43 gap junctions is an independent marker for breast tumors. Cancer Res. 59:4104–4110.[Abstract/Free Full Text]
  7. Qin H., Shao Q., Curtis H., Galipeau J., Belliveau D.J., Wang T., Alaoui-Jamali M.A., Laird D.W. (2002) Retroviral delivery of connexin genes to human breast tumor cells inhibits in vivo tumor growth by a mechanism that is independent of significant gap junctional intercellular communication. J. Biol. Chem. 277:29132–29138.[Abstract/Free Full Text]
  8. Qin H., Shao Q., Thomas T., Kalra J., Alaoui-Jamali M.A., Laird D.W. (2003) Connexin26 regulates the expression of angiogenesis-related genes in human breast tumor cells by both GJIC-dependent and -independent mechanisms. Cell Commun. Adhes. 10:387–393.[ISI][Medline]
  9. Monaghan P., Clarke C., Perusinghe N.P., Moss D.W., Chen X.Y., Evans W.H. (1996) Gap junction distribution and connexin expression in human breast. Exper. Cell Res. 223:29–38.[CrossRef][ISI][Medline]
  10. Monaghan P. and Moss D. (1996) Connexin expression and gap junctions in the mammary gland. Cell Biol. Int. 20:121–125.[CrossRef][ISI][Medline]
  11. Locke D. (1998) Gap junctions in normal and neoplastic mammary gland. J. Pathol. 186:343–349.[CrossRef][ISI][Medline]
  12. Kang K.S., Wilson M.R., Hayashi T., Chang C.C., Trosko J.E. (1996) Inhibition of gap junctional intercellular communication in normal human breast epithelial cells after treatment with pesticides, PCBs, and PBBs, alone or in mixtures. Environ. Health Perspect. 104:192–200.[ISI][Medline]
  13. Lee S.W., Tomasetto C., Sager R. (1991) Positive selection of candidate tumor-suppressor genes by subtractive hybridization. Proc. Natl Acad. Sci. USA 88:2825–2829.[Abstract/Free Full Text]
  14. Lee S.W., Tomasetto C., Paul D., Keyomarsi K., Sager R. (1992) Transcriptional downregulation of gap-junction proteins blocks junctional communication in human mammary tumor cell lines. J. Cell. Biol. 118:1213–1221.[Abstract/Free Full Text]
  15. Tomasetto C., Neveu M.J., Daley J., Horan P.K., Sager R. (1993) Specificity of gap junction communication among human mammary cells and connexin transfectants in culture. J. Cell. Biol. 122:157–167.[Abstract/Free Full Text]
  16. Hirschi K.K., Xu C.E., Tsukamoto T., Sager R. (1996) Gap junction genes Cx26 and Cx43 individually suppress the cancer phenotype of human mammary carcinoma cells and restore differentiation potential. Cell Growth Differ. 7:861–870.[Abstract]
  17. Momiyama M., Omori Y., Ishizaki Y., Nishikawa Y., Tokairin T., Ogawa J., Enomoto K. (2003) Connexin26-mediated gap junctional communication reverses the malignant phenotype of MCF-7 breast cancer cells. Cancer Sci. 94:501–507.[CrossRef]
  18. Goodenough D.A. and Paul D.L. (2003) Beyond the gap: functions of unpaired connexon channels. Nat. Rev. Mol. Cell Biol. 4:285–294.[CrossRef][ISI][Medline]
  19. Meyer R.A., Laird D.W., Revel J.P., Johnson R.G. (1992) Inhibition of gap junction and adherens junction assembly by connexin and A-CAM antibodies. J. Cell. Biol. 119:179–189.[Abstract/Free Full Text]
  20. Kojima T., Spray D.C., Kokai Y., Chiba H., Mochizuki Y., Sawada N. (2002) Cx32 formation and/or Cx32-mediated intercellular communication induces expression and function of tight junctions in hepatocytic cell line. Exp. Cell Res. 276:40–51.[CrossRef][ISI][Medline]
  21. Duffy H.S., John G.R., Lee S.C., Brosnan C.F., Spray D.C. (2000) Reciprocal regulation of the junctional proteins claudin-1 and connexin43 by interleukin-1beta in primary human fetal astrocytes. J. Neurosci. 20:RC114.[Abstract/Free Full Text]
  22. Kausalya P.J., Phua D.C., Hunziker W. (2004) Association of ARVCF with zonula occludens (ZO)-1 and ZO-2: binding to PDZ-domain proteins and cell-cell adhesion regulate plasma membrane and nuclear localization of ARVCF. Mol. Biol. Cell 15:5503–5515.[Abstract/Free Full Text]
  23. Herve J.C., Bourmeyster N., Sarrouilhe D. (2004) Diversity in protein-protein interactions of connexins: emerging roles. Biochim. Biophys. Acta 1662:22–41.[Medline]
  24. Giepmans B.N. (2004) Gap junctions and connexin-interacting proteins. Cardiovasc. Res. 62:233–245.[Abstract/Free Full Text]
  25. Laird D.W. (2006) Life cycle of connexins in health and disease. Biochem. J. 394:527–543.[CrossRef][ISI][Medline]
  26. Duflot-Dancer A., Mesnil M., Yamasaki H. (1997) Dominant-negative abrogation of connexin-mediated cell growth control by mutant connexin genes. Oncogene 15:2151–2158.[CrossRef][ISI][Medline]
  27. Yamasaki H., Krutovskikh V., Mesnil M., Tanaka T., Zaidan-Dagli M.L., Omori Y. (1999) Role of connexin (gap junction) genes in cell growth control and carcinogenesis. Comptes Rendus de l Academie des Sciences Serie Iii, Sciences de la Vie 322:151–159.[CrossRef]
  28. Krutovskikh V.A., Troyanovsky S.M., Piccoli C., Tsuda H., Asamoto M., Yamasaki H. (2000) Differential effect of subcellular localization of communication impairing gap junction protein connexin43 on tumor cell growth in vivo. Oncogene 19:505–513.[CrossRef][ISI][Medline]
  29. Zhang Y.W., Kaneda M., Morita I. (2003) The gap junction-independent tumor-suppressing effect of connexin 43. J. Biol. Chem. 278:44852–44856.[Abstract/Free Full Text]
  30. Fu C.T., Bechberger J.F., Ozog M.A., Perbal B., Naus C.C. (2004) CCN3 (NOV) interacts with connexin43 in C6 glioma cells: possible mechanism of connexin-mediated growth suppression. J. Biol. Chem. 279:36943–36950.[Abstract/Free Full Text]
  31. Zhang Y.W., Nakayama K., Morita I. (2003) A novel route for connexin 43 to inhibit cell proliferation: negative regulation of S-phase kinase-associated protein (Skp 2). Cancer Res. 63:1623–1630.[Abstract/Free Full Text]
  32. el-Sabban M.E. and Pauli B.U. (1994) Adhesion-mediated gap junctional communication between lung-metastatatic cancer cells and endothelium. Invas. Metastasis 14:164–176.
  33. Kamibayashi Y., Oyamada Y., Mori M., Oyamada M. (1995) Aberrant expression of gap junction proteins (connexins) is associated with tumor progression during multistage mouse skin carcinogenesis in vivo. Carcinogenesis 16:1287–1297.[Abstract/Free Full Text]
  34. Saunders M.M., Seraj M.J., Li Z., Zhou Z., Winter C.R., Welch D.R., Donahue H.J. (2001) Breast cancer metastatic potential correlates with a breakdown in homospecific and heterospecific gap junctional intercellular communication. Cancer Res. 61:1765–1767.[Abstract/Free Full Text]
  35. Hamada J., Nagayasu H., Kawano T., Mizutani T., Nakata D., Hosokawa M., Takeichi N. (1998) Characterization of the progressive sublines derived from a weakly malignant cloned cell line, ER-1, co-inoculated subcutaneously with a foreign body. Clin. Exp. Metastasis 16:291–298.[CrossRef][ISI][Medline]
  36. Shoji A., Sakamoto Y., Tsuchiya T., Moriyama K., Kaneko T., Okubo T., Umeda M., Miyazaki K. (1997) Inhibition of tumor promoter activity toward mouse fibroblasts and their in vitro transformation by tissue inhibitor of metalloproteinases-1 (TIMP-1). Carcinogenesis 18:2093–2100.[Abstract/Free Full Text]
  37. El-Sabban M.E. and Pauli B.U. (1991) Cytoplasmic dye transfer between metastatic tumor cells and vascular endothelium. J. Cell Biol. 115:1375–1382.[Abstract/Free Full Text]
  38. Ito A., Katoh F., Kataoka T.R., et al. (2000) A role for heterologous gap junctions between melanoma and endothelial cells in metastasis. J. Clin. Invest. 105:1189–1197.[ISI][Medline]
  39. Qin H., Shao Q., Igdoura S.A., Alaoui-Jamali M.A., Laird D.W. (2003) Lysosomal and proteasomal degradation play distinct roles in the life cycle of Cx43 in gap junctional intercellular communication-deficient and -competent breast tumor cells. J. Biol. Chem. 278:30005–30014.[Abstract/Free Full Text]
  40. Thomas T., Telford D., Laird D.W. (2004) Functional domain mapping and selective trans-dominant effects exhibited by Cx26 disease-causing mutations. J. Biol. Chem. 279:19157–19168.[Abstract/Free Full Text]
  41. Mao A.J., Bechberger J., Lidington D., Galipeau J., Laird D.W., Naus C.C. (2000) Neuronal differentiation and growth control of neuro-2a cells after retroviral gene delivery of connexin43. J. Biol. Chem. 275:34407–34414.[Abstract/Free Full Text]
  42. Laird D.W., Castillo M., Kasprzak L. (1995) Gap junction turnover, intracellular trafficking, and phosphorylation of connexin43 in brefeldin A-treated rat mammary tumor cells. J. Cell Biol. 131:1193–1203.[Abstract/Free Full Text]
  43. Yeow K.M., Phillips B.W., Beaudry P.P., Leco K.J., Murphy G., Edwards D.R. (2001) Expression of MMPs and TIMPs in mammalian cells. Methods Mol. Biol. 151:181–189.[Medline]
  44. Fata J.E., Leco K.J., Moorehead R.A., Martin D.C., Khokha R. (1999) Timp-1 is important for epithelial proliferation and branching morphogenesis during mouse mammary development. Dev. Biol. 211:238–254.[CrossRef][ISI][Medline]
  45. Locke D., Perusinghe N., Newman T., Jayatilake H., Evans W.H., Monaghan P. (2000) Developmental expression and assembly of connexins into homomeric and heteromeric gap junction hemichannels in the mouse mammary gland. J. Cell. Physiol. 183:228–237.[CrossRef][ISI][Medline]
  46. Laird D.W., Jordan K., Shao Q. (2001) Expression and imaging of connexin-GFP chimeras in live mammalian cells. Methods Mol. Biol. 154:135–142.[Medline]
  47. Contreras J.E., Saez J.C., Bukauskas F.F., Bennett M.V. (2003) Functioning of cx43 hemichannels demonstrated by single channel properties. Cell Commun. Adhes. 10:245–249.[ISI][Medline]
  48. Hernandez-Blazquez F.J., Joazeiro P.P., Omori Y., Yamasaki H. (2001) Control of intracellular movement of connexins by E-cadherin in murine skin papilloma cells. Exp. Cell Res. 270:235–247.[CrossRef][ISI][Medline]
  49. Duffy H.S., Delmar M., Spray D.C. (2002) Formation of the gap junction nexus: binding partners for connexins. J. Physiol. (Paris) 96:243–249.[CrossRef][ISI][Medline]
  50. Giancotti F.G. and Ruoslahti E. (1999) Integrin signaling. Science 285:1028–1032.[Abstract/Free Full Text]
  51. Brakebusch C. and Fassler R. (2003) The integrin-actin connection, an eternal love affair. EMBO J. 22:2324–2333.[CrossRef][ISI][Medline]
  52. Tate A.W., Lung T., Radhakrishnan A., Lim S.D., Lin X., Edlund M. (2006) Changes in gap junctional connexin isoforms during prostate cancer progression. Prostate 66:19–31.[CrossRef][ISI][Medline]
  53. El-Sabban M.E., Sfeir A.J., Daher M.H., Kalaany N.Y., Bassam R.A., Talhouk R.S. (2003) ECM-induced gap junctional communication enhances mammary epithelial cell differentiation. J. Cell Sci. 116:3531–3541.[Abstract/Free Full Text]
  54. White D.E., Kurpios N.A., Zuo D., Hassell J.A., Blaess S., Mueller U., Muller W.J. (2004) Targeted disruption of beta1-integrin in a transgenic mouse model of human breast cancer reveals an essential role in mammary tumor induction. Cancer Cell 6:159–170.[CrossRef][ISI][Medline]
  55. Stetler-Stevenson W.G., Hewitt R., Corcoran M. (1996) Matrix metalloproteinases and tumor invasion: from correlation and causality to the clinic. Semin. Cancer Biol. 7:147–154.[CrossRef][ISI][Medline]
  56. Kugler A. (1999) Matrix metalloproteinases and their inhibitors. Anticancer Res. 19:1589–1592.[ISI][Medline]
  57. Bartsch J.E., Staren E.D., Appert H.E. (2003) Matrix metalloproteinase expression in breast cancer. J. Surg. Res. 110:383–392.[CrossRef][ISI][Medline]
  58. Morini M., Mottolese M., Ferrari N., Ghiorzo F., Buglioni S., Mortarini R., Noonan D.M., Natali P.G., Albini A. (2000) The alpha 3 beta 1 integrin is associated with mammary carcinoma cell metastasis, invasion, and gelatinase B (MMP-9) activity. Int. J. Cancer 87:336–342.[CrossRef][ISI][Medline]
  59. Mesnil M., Krutovskikh V., Piccoli C., Elfgang C., Traub O., Willecke K., Yamasaki H. (1995) Negative growth control of HeLa cells by connexin genes: connexin species specificity. Cancer Res. 55:629–639.[Abstract/Free Full Text]
  60. Henzl M.T., Thalmann I., Larson J.D., Ignatova E.G., Thalmann R. (2004) The cochlear F-box protein OCP1 associates with OCP2 and connexin 26. Hear. Res. 191:101–109.[CrossRef][ISI][Medline]
Received December 20, 2005; revised May 18, 2006; accepted May 27, 2006.


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
27/12/2528    most recent
bgl110v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (6)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Kalra, J.
Right arrow Articles by Laird, D. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kalra, J.
Right arrow Articles by Laird, D. W.
Social Bookmarking