Advertisement
JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M609653200 on April 9, 2007

J. Biol. Chem., Vol. 282, Issue 26, 19190-19202, June 29, 2007
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
282/26/19190    most recent
M609653200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gong, X.-Q.
Right arrow Articles by Laird, D. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gong, X.-Q.
Right arrow Articles by Laird, D. W.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Differential Potency of Dominant Negative Connexin43 Mutants in Oculodentodigital Dysplasia*

Xiang-Qun Gong{ddagger}, Qing Shao§, Stéphanie Langlois§, Donglin Bai{ddagger}21, and Dale W. Laird{ddagger}§2

From the Departments of {ddagger}Physiology and Pharmacology and of §Anatomy and Cell Biology, University of Western Ontario, London, Ontario N6A 5C1, Canada

Received for publication, October 13, 2006 , and in revised form, April 3, 2007.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Oculodentodigital dysplasia (ODDD) is a congenital autosomal dominant disorder with phenotypic variability, which has been associated with mutations in the GJA1 gene encoding connexin43 (Cx43). Given that Cx43 mutants are thought to be equally co-expressed with wild-type Cx43 in ODDD patients, it is imperative to examine the consequence of these mutants in model systems that reflect this molar ratio. To that end, we used differential fluorescent protein tagging of mutant and wild-type Cx43 to quantitatively monitor the ratio of mutant/wild-type within the same putative gap junction plaques and co-immunoprecipitation to determine if the mutants interact with wild-type Cx43. Together the fluorescence-based assay was combined with patch clamp analysis to assess the dominant negative potency of Cx43 mutants. Our results revealed that the ODDD-linked Cx43 mutants, G21R and G138R, as well as amino terminus green fluorescent protein-tagged Cx43, were able to co-localize with wild-type Cx43 at the gap junction plaque-like structures and to co-immunoprecipitate with wild-type Cx43. All Cx43 mutants demonstrated dominant negative action on gap junctional conductance of wild-type Cx43 but not that of Cx32. More interestingly, these Cx43 mutants demonstrated different potencies in inhibiting the function of wild-type Cx43 with the G21R mutant being two times more potent than the G138R mutant. The potency difference in the dominant negative properties of ODDD-linked Cx43 mutants may have clinical implications for the various symptoms and disease severity observed in ODDD patients.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Gap junctions (GJ)3 are membrane channels that allow direct cell-to-cell transfer of ions, signaling molecules and metabolites of <1 kDa in size. Gap junction proteins, connexins, oligomerize to form connexons (hemichannels) that dock with connexons from the contacting cell to form gap junction channels (1). At the cell-cell interface, hundreds to thousands of gap junction channels cluster to form gap junction plaques (2). There are 21 members in the human connexin family, all of which share a common membrane topology. Every connexin consists of four transmembrane domains (M1 to M4), two extracellular loops (EL-1 and EL-2), and one intracellular loop (IL) with both the amino terminus (AT) and carboxyl terminus (CT) exposed to the cytoplasm (3-6).

Connexins are expressed in a tissue- and developmental stage-dependent manner, and many cells express multiple connexin isoforms. Connexins of the same kind oligomerize to form homomeric hemichannels, whereas oligomerization of compatible connexins of different isoforms can form heteromeric hemichannels. Moreover, both homotypic and heterotypic intercellular gap junction channels exist in native tissues (7, 8). This diversity in channel composition reflects the complex nature of gap junctional intercellular communication (GJIC), which in turn, plays important roles in many physiological and developmental processes (2).

The expression of dominant negative variants of proteins has been successfully used to elucidate protein function, not only within the context of composite cellular and tissue organizations, but also within protein complexes (9, 10). Given the oligomeric nature of gap junction channels, a mutant connexin subunit that is still capable of oligomerization might exert dominant negative effects on GJIC, if it were to associate with wild-type gap junction proteins. Hence, disease-linked mutant connexins are ideal candidates for exerting dominant negative effects on compatible wild-type connexins. Interestingly, many diseases linked to connexin mutations are autosomal dominant leading to speculation that the mutant encoded by one allele is dominant to the co-expressed wild-type connexin (11-13).

One newly characterized connexin-linked human disease is oculodentodigital dysplasia (ODDD). ODDD is a rare inherited disorder affecting the development of the face, teeth, eyes, and limbs, where patients display symptoms of congenital craniofacial deformities and limb abnormalities (14). The majority of the ODDD cases are inherited in an autosomal dominant fashion (15). ODDD has been associated with over 30 known human Cx43 mutations. Thus far, functional studies have been performed on 12 ODDD-linked Cx43 mutants, all of which demonstrated functionally impaired gap junctions when expressed in GJIC-incompetent N2A and/or HeLa cells (16-20). However, different mechanisms appear to be responsible for the loss of Cx43 function, as some mutants display altered trafficking while others show no apparent trafficking defect but exhibit either a complete or partial loss of gap junction channel or hemichannel function (16-21). Interestingly, many ODDD-linked mutants, including G21R and G138R, have been shown to exhibit dominant negative properties on gap junction function in cell lines that express endogenous Cx43 (18, 19). Our partial characterization of these two ODDD-linked mutants in overexpression models failed to see any difference between G21R and G138R, in terms of trafficking, the formation of gap junction plaque-like structures or their functional status (18). On the other hand, the patients carrying these two mutants not only displayed several common ODDD symptoms such as digit syndactyly and facial anomalies but also distinct clinical defects possibly reflecting differences that could be attributed to the site of the mutation. For example, a patient with the G21R mutation displayed an atrial-septal defect in the heart, whereas patients harboring the G138R mutation exhibited combinations of neurological disorders, hearing loss, spastic gait, and bladder disorders (14, 22, 23). Based on these clinical observations, we hypothesized that different Cx43 mutants may in fact have different potentials at inhibiting the function of co-expressed wild-type Cx43.

To test our hypothesis at the molecular level, it was necessary to establish a differential fluorescent protein tagging strategy where wild-type Cx43 was tagged with monomeric red fluorescent protein (Cx43-mRFP), and Cx43 mutants were tagged with green fluorescent protein (GFP). In addition, to establish a proof of principle and determine the efficacy of using two distinct fluorescent protein tags, we assessed the effect of a loss-of-function mutant where GFP was tagged to the amino-terminal of Cx43 (GFP-Cx43) on Cx43-mRFP. Fluorescent protein-tagged connexins, where the fluorescent protein is tagged to the carboxyl terminus of the connexin (Cxs-GFP), have been widely used in elucidating the pathways and events linked to connexin assembly, trafficking, localization, function, and degradation (24-30). Importantly, these fusion proteins exhibit characteristics similar to their wild-type connexin counterparts. For example, when expressed in various cell lines, Cx43-GFP, Cx32-GFP, and Cx26-GFP were all able to traffic to the cell surface and form functional gap junction channels and even propagate calcium waves (24, 26, 27, 29). Likewise, patch clamp recordings revealed that Cx43-GFP channels exhibit similar macroscopic and single channel conductance as those of untagged Cx43, although they are less sensitive to transjunctional voltage and lack fast voltage gating to a residual state (25, 28, 30). On the other hand, GFP-Cx43 was found to be incapable of forming functional gap junction channels or hemichannels while still being assembled into gap junction-like structures at the cell surface (29, 30).

In this study we combined differential fluorescent protein tagging, co-immunoprecipitation experiments, and electrophysiological recordings to investigate the mechanism of dominant negative action of ODDD-linked Cx43 mutants in a dosage control environment. Our data indicate that the ODDD-linked mutants, G21R and G138R, were both dominant to wild-type Cx43 when expressed at a 1:1 ratio, whereas the G21R mutant was clearly more potent. Collectively, these data support a model where patients harboring the G21R mutant would exhibit less overall Cx43-based GJIC than patients carrying the G138R mutant possibly explaining the differences in overall disease burden.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Cell Culture and Reagents—All media, sera, and culture reagents were obtained from Invitrogen, BD Biosciences, or Sigma. N2A (mouse neuroblastoma) cells, normal rat kidney (NRK) cells, HeLa (human cervical carcinoma) cells and BICR-M1Rk (rat mammary) tumor cells were cultured as we previously reported (24, 31). The HeLa-Cx32D cell line representing HeLa cells stably expressing Cx32 was kindly provided by Dr. Michael Koval (University of Pennsylvania School of Medicine, Philadelphia, PA).

Engineering of Connexin-chimeric and Mutant Cx43 cDNAs—The vector encoding Cx43-GFP used in this study was constructed as described in Jordan et al. (24) (Fig. 1B). The engineering of the GFP-Cx43 and Cx32-GFP constructs was described previously (29) (Fig. 1, A and D).

To generate the GFP-Cx32 fusion protein, the cDNA of rat Cx32 was amplified by PCR with the forward primer of Cx32 (5'-GCC CTC GAG ATG AAC TGG ACA) to introduce the XhoI site and the reverse primer (5'-GCG GAA TTC GTT AGC AGG) to introduce the EcoRI site. PCR products were digested with XhoI/EcoRI and cloned into the XhoI/EcoRI-cut pEGFP-C3 (Clontech) vector. DNA sequence analysis confirmed that Cx32 was fused in-frame to the carboxyl terminus of GFP with the addition of an amino acid linker encoded by TAC TCA GAT CTC GAG (Fig. 1C).

Two human ODDD Cx43 mutants, G21R and G138R, were constructed using the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA) as previously described (18). The monomeric mRFP (variant of DsRed) tagged to the carboxyl terminus of Cx43 (Cx43-mRFP) was kindly provided by Dr. Guido Gaietta (University of California, San Diego, CA) (32).

Transient and Stable Transfection—Cells were transfected with GFP- or mRFP-tagged Cxs by Lipofectamine PLUS or Lipofectamine 2000 (Invitrogen) as recommended by the manufacturer (as described in Refs. 18 and 20). In co-transfection experiments, cDNAs of Cx43 mutants (or GFP-Cx43) and Cx43-mRFP were mixed in designed ratios, whereas Cx43-mRFP cDNA was kept constant at 3 µg (20). Because the mRFP sequence was only 14 amino acid residues shorter than the GFP sequence, the calculated ratios of cDNAs used in transfection studies should only be minimally affected. 24-48 h after transient transfection, cells were used for patch clamp recording, microinjection, or fixed for immunocytochemistry. For stable transfection, cells were sub-cultured into selection medium containing previously titrated concentrations of G418, and cell clones were selected as described previously (24).

Immunocytochemistry—Cells were immunolabeled as previously described (33). Briefly, cells were grown on glass cover-slips and fixed with 80% methanol/20% acetone at 4 °C for 15 min. Cells expressing Cx43 or GFP-Cx43 were labeled with a 1:50 dilution of anti-Cx43 monoclonal antibody or polyclonal antibody (specific for the 360-382 peptide segment of Cx43) (33). Cx32-, GFP-Cx32-, and Cx32-GFP-expressing cells were labeled with 1:2000 dilution of an anti-Cx32 (Ham8) monoclonal antibody, specific for an 11-peptide sequence (229-239) located in the cytoplasmic region of Cx32 (34). The Ham8 antibody was generously provided by Dr. Fujikura (Yamaguchi University School of Medicine, Yamaguchi, Japan). Washed cells were incubated for 1 h in goat anti-mouse or donkey anti-rabbit antibodies conjugated to Texas red or fluorescein isothiocyanate (Jackson ImmunoResearch Laboratories, Inc., PA). Coverslips were rinsed in distilled water, mounted, and analyzed as previously described (33).


Figure 1
View larger version (43K):
[in this window]
[in a new window]

  		FIGURE 1.
Schematic models of GFP-tagged connexins and the amino acid locations of the ODDD-linked mutations, G21R and G138R. In GFP-Cx43 (A), 18 amino acids link GFP to the amino terminus (AT) of Cx43, and the last 2 amino acid residues at carboxyl terminus (CT) are replaced with 7 nonsense amino acid residues remaining from the Cx43-GFP construct from which it was derived. GFP was tagged to the carboxyl terminus of Cx43 via a 7-amino acid linker to form Cx43-GFP (B). In GFP-Cx32, the linker joining GFP to Cx32 consisted of 5 amino acid residues (C), and in Cx32-GFP, 7 amino acid residues link GFP to the carboxyl terminus of Cx32 (D). The human ODDD mutation, G21R, is located at the beginning of the M1 domain of Cx43, and the G138R mutation is located within the intracellular loop (IL). Both mutants are shown in panel B (arrows). EL-1 and EL-2 represent the two extracellular loops, and M1 to M4 represent the four transmembrane domains.

 
Confocal Imaging—All cells expressing GFP-tagged Cxs/mutants or Cx43-mRFP, and cells used for antibody labeling, were fixed as described above and imaged on a Zeiss (Thornwood, NY, USA) LSM 510 META confocal microscope mounted on an inverted Axiovert 200 motorized microscope equipped with a 63x oil (1.4 numerical aperture) objective. Fluorescence signals were imaged after excitation with 488, 543, or 730 nm laser lines produced by argon, heliumneon, or tunable Chameleon multiphoton lasers, respectively. All fluorescence signals were collected on a photomultiplier after a passage through appropriate filter sets. In quantitative analyses, all 12-bit images were taken with identical settings. The regions of interest (ROIs) were carefully selected to include putative gap junctional plaques at cell-cell interfaces. The accuracy of the selected ROIs was double checked at high magnification to ensure that the plaque was properly demarcated, whereas fluorescence intensity measurements were collected from the original images. Based on the fluorescence intensity histograms generated from the ROIs, mean fluorescence densities from both red (mRFP) and green (GFP) channels were determined, and the green/red ratio was calculated. Digital images were prepared using Zeiss LSM, Adobe Photoshop 7.0, and CorelDraw 12 software (31). Expression of GFP-tagged connexins in HeLa cells promotes the formation of oversized gap junctions unlike most gap junctions observed in vivo, which tend to be much smaller. However, these large gap junctions offer a convenient experimental model for imaging and to define regions of interest for our quantitative analysis of connexin intermixing within the plaque, whereas limitations in resolution prevent any assignment of connexins to specific sides of the gap junction plaque.

Fluorescent Dye Microinjection—Control cells or cell clusters expressing exogenous fluorescent protein-tagged connexins with clear fluorescent plaques at sites of cell-cell apposition were selected for microinjection. Cells were pressure microinjected with 5% Lucifer yellow (Molecular Probes, Leiden, Netherlands) using an Eppendorf FemtoJet automated pressure microinjector and a Leica DM IRE2 inverted epifluorescence microscope. Approximately, 1-4 min after microinjection, digital images were collected with a charge-coupled device camera (Hamamatsu Photonics, Japan) using OpenLab software (Improvision). The percentage of microinjected cells that exhibited dye coupling was determined as previously described (18).


Figure 2
View larger version (72K):
[in this window]
[in a new window]

  		FIGURE 2.
GFP-Cx32 and GFP-Cx43 assemble into gap junctional plaque-like structures at cell-cell interfaces of BICR-M1RK and HeLa cells. BICR-M1RK cells transfected with cDNAs encoding GFP-Cx43 (A) or GFP-Cx32 (B) were fixed and immunolabeled for Cx43 (anti-Cx43) to assess the distribution pattern of the GFP-tagged connexins with respect to endogenous Cx43. Connexin-deficient HeLa cells were also engineered to express GFP-Cx43 (C) or GFP-Cx32 (D) and antibodies to Cx43 (anti-Cx43) or Cx32 (anti-Cx32) were used to immunostain the expressed connexin proteins. High resolution confocal microscopy revealed that GFP-Cx32 (F) and immunolabeled endogenous Cx43 (E) co-localize to a sub-population of both intracellular transport intermediates and gap junction plaques (G) in BICR-M1RK cells. Although GFP-Cx32 and Cx43 appear to co-exist in some plaques (H; arrows), due to resolution limits, Cx43 and Cx32 could also be differentially localized to adjacent plaques or subdomains within plaques (H; Cx32, open arrowheads; Cx43, line arrowheads). Bar = 10 µm.

 
Immunoprecipitation and Western Blotting—For the co-immunoprecipitation studies, HeLa cells were transfected with Cx43-mRFP together with Cx32-GFP or Cx43 variants (Cx43-GFP, G21R-GFP, G138R-GFP, or GFP-Cx43). Identical amounts of transfected HeLa cells were incubated in the immunoprecipitation (IP) lysis buffer (150 mM NaCl, 10 mM Tris-HCl, pH 7.4, 1 mM EDTA, 1 mM EGTA, 0.5% Nonidet P-40, 1% Triton X-100, 1 mM sodium orthovanadate, 1 mM sodium fluoride, and proteases inhibitors) overnight at 4 °C in the presence of an anti-GFP antibody. Immune complexes were collected by incubating the mixtures with 30 µl (50% suspension) of protein A-Sepharose beads. Nonspecifically bound proteins were removed by washing the beads three times in 500 µl of IP lysis buffer. Bound material was solubilized in 30 µl of 2-fold concentrated Laemmli sample buffer, boiled for 5 min, and resolved by SDS-PAGE. The proteins were transferred onto nitrocellulose membranes, blocked 30 min with the Odyssey (Li-Cor Biosciences) blocking buffer at room temperature, and incubated with primary antibody (GFP monoclonal antibody (MAB3580, Chemicon, Temecula, CA) or DsRed (mRFP) polyclonal antibody (632496, Clontech, Mountain View, CA)) overnight at 4 °C. Immunoreactive bands were revealed following 1-h incubation with Alexa Fluor 680 (red, Invitrogen) or IRDye 800 conjugated secondary antibodies (green, Rockland, Gilbertsville, PA) and visualized using the Odyssey Infrared Imaging System (Li-Cor Biosciences, Lincoln, NE).

Patch Clamp Electrophysiology—Functional gap junction coupling between paired N2A, BICR-M1Rk, and HeLa-Cx32D cells expressing fluorescent protein-tagged mutant and/or wild-type Cx43 and Cx32 was assessed using the dual whole cell patch clamp technique as previously described (18, 20). N2A cells were chosen due to the facts that: 1) they are GJIC-deficient; 2) they can be easily transfected with different connexin encoding cDNAs; and 3) they are readily amenable to the patch clamp technique. Isolated cell pairs with green fluorescent plaques (representing GFP-tagged Cx) at cell-cell contacts or with both green and red (representing GFP and mRFP-tagged Cx or mutants, respectively) plaques (in dual-color experiments) at the cell-cell interfaces were chosen for patch clamp recording. Gap junctional conductance (Gj) was determined and presented as mean ± S.E. Online series resistance compensation at 80% or off-line series resistance compensation (35) were applied to improve the accuracy of measured Gj. Student's t test was performed with Microsoft Excel. The dose-inhibition curves were fitted with GraphPad Software Prism version 4.


Figure 3
View larger version (22K):
[in this window]
[in a new window]

  		FIGURE 3.
GFP-Cx32 and GFP-Cx43 selectively inhibit the gap junction conductance of their wild-type connexin counterparts. BICR-M1RK and HeLa-Cx32D cells expressing Cx43 and Cx32, respectively, were transfected with cDNAs encoding GFP-Cx32, GFP-Cx43, Cx32-GFP, and Cx43-GFP. Isolated BICR-M1RK or HeLa-Cx32D cell pairs with green plaque-like structures at cell-cell interfaces were used for dual whole cell patch clamp recording to measure gap junction conductance (A and B). Data were normalized to control and presented as mean ± S.E., after compensation the averaged intercellular coupling conductances were 75 nS for BICR-M1Rk and 65 nS for HeLa-Cx32D cell pairs. 10 ≤ n ≤ 21 for each data point. **, p < 0.01; ***, p < 0.001. NRK, BICR-M1RK, and HeLa-Cx32D cell clusters expressing GFP-Cx43 or GFP-Cx32 were selected for Lucifer yellow microinjection and dye transfer analysis. A summary of the dye transfer data was plotted as a bar graph (C). As controls, all three cell lines showed excellent dye coupling. Whereas dye spread extensively in NRK and BICR-M1RK cells expressing GFP-Cx32, HeLa-Cx32D cells expressing GFP-Cx32 exhibited dramatically reduced dye coupling. Conversely, the expression of GFP-Cx43 resulted in a reduction of >97%in the incidence of dye transfer in NRK and BICR-M1RK cells but failed to significantly decrease dye coupling in HeLa-Cx32D cells (C). The numbers in parentheses represent the number of injected cells for each experimental condition.

 

   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
To determine if specific Cx43 ODDD-linked mutants exhibit the same potential to impair co-expressed Cx43 function, it was essential to recapitulate an in vitro model that would mimic the human disease where both mutant and endogenous Cx43 are expected to be equally co-expressed. To that end, we established a correlative two-color fluorescent protein tagging and electrophysiological recordings model under conditions where connexin variants were co-expressed within the same cells at different dose ratios.

Characterization of Connexin Variants and Establishment of a Correlative Two-color Fluorescence and Electrophysiological Assay—To begin to assess the specificity of connexin variants to act as specific dominant negatives we first characterized the spatial localization of GFP-Cx32 in Cx43-positive BICR-M1Rk cells and GJIC-deficient HeLa cells. GFP-Cx32 formed gap junction plaque-like structures with or without the presence of endogenous Cx43 (Fig. 2, B and D) not unlike GFP-Cx43 (Fig. 2, A and C). To assess if GFP-Cx32 and endogenous Cx43 were co-localized to the same intracellular compartments and cell-cell junctions, confocal microscopy was used to examine the distribution of GFP-Cx32 in BICR-M1Rk cells immunolabeled for Cx43 (Fig. 2, E-H). This study revealed that GFP-Cx32 and endogenous Cx43 distributed to both unique and common intracellular vesicles and/or compartments. Interestingly, at the cell surface GFP-Cx32 appeared to partially co-localized to the same gap junction plague-like structures as Cx43 (Fig. 2H, arrows), whereas other putative Cx43 gap junction plaques were essentially devoid of GFP-Cx32 or vice versa (Fig. 2H, Cx43, line arrowheads; GFP-Cx32, open arrowheads). It is important to note that in cases where Cx43 and Cx32 appear to occupy the same plaque they could be sequestered into separate domains of the gap junction plaque or be confined to two closely apposed plaques.

To examine whether the GFP-Cx32 is able to form functional gap junction channels, both Lucifer yellow microinjection and dual whole cell patch clamp techniques were used to assess the functional status of GFP-Cx32 in GJIC-deficient N2A cells. When transfected with GFP-Cx32, most (79/82) of the GFP-Cx32-expressing cells failed to pass Lucifer yellow dye to neighboring cells, and the vast majority (18/20) of the cell pairs demonstrated no electrical coupling (Table 1). Only 2/20 cells displayed a low level of coupling conductance that was comparable to that observed for non-transfected N2A cells (36). Likewise, in the case of GFP-Cx43 only minimal electrical coupling was detected with no evidence of dye coupling (Table 1). In contrast, Cx32-GFP and Cx43-GFP not only formed fluorescent gap junction plaques but also exhibited robust coupling (Table 1).


View this table:
[in this window]
[in a new window]

  		TABLE 1
GFP-Cx32 and GFP-Cx43 do not form functional gap junction channels in N2A cells

N2A cells were transfected with cDNAs encoding GFP-Cx43, GFP-Cx32, Cx43-GFP, and Cx32-GFP, and GJIC was assessed by dual whole cell patch clamp and dye coupling assays. Isolated cell pairs that exhibited green fluorescent plaques at the cell-cell interfaces were chosen as candidates for electrophysiological recording or Lucifer yellow microinjection.

 
To assess if non-functional connexin variants could specifically inhibit their functional counterparts in a specific manner, we expressed various connexin variants in cells that expressed either endogenous Cx43 (BICR-M1Rk) or HeLa cells engineered to express Cx32. Isolated cell pairs that had GFP-positive gap junction plaque-like structures at cell-cell interfaces were selected for dual whole cell patch clamp recordings and the junctional conductance was determined. In BICR-M1Rk cells, which contain abundant endogenous Cx43, the expression of GFP-Cx43 significantly reduced cell-cell coupling by ~50%, whereas GFP-Cx32 had no effect (Fig. 3A). Also as expected, Cx43-GFP and Cx32-GFP did not reduce the coupling level of BICR-M1Rk cells. In HeLa-Cx32D cells, which stably express Cx32, GFP-Cx32 expression dramatically reduced intercellular coupling conductance by nearly 70%, whereas GFP-Cx43, Cx32-GFP, and Cx43-GFP did not reduce GJIC (Fig. 3B). These findings were confirmed in dye coupling assays where GFP-Cx32 was specifically found to reduce dye coupling in HeLa-Cx32D cells and GFP-Cx43 was found to reduce dye coupling in both Cx43-positive NRK and BICR-M1Rk cells (Fig. 3C). Collectively, these results demonstrated that GFP-Cx32 and GFP-Cx43 selectively and robustly inhibit only the GJIC-mediated by their corresponding wild-type connexins, in a dominant negative manner.


Figure 4
View larger version (44K):
[in this window]
[in a new window]

  		FIGURE 4.
GFP-Cx43 inhibited Cx43-mediated GJIC in a dose-dependent manner. GJIC-deficient HeLa and N2A cells were transfected with mixed cDNAs encoding GFP-Cx43 and Cx43-mRFP at various ratios. A constant amount (3 µg) of Cx43-mRFP cDNA was used, whereas GFP-Cx43 cDNA was increased from 0 to 15 µg as shown at the bottom of panel C. Individual cell pairs with green and red plaques co-localized at cell-cell interfaces for N2A cells (B) were selected for dual whole cell patch clamp recordings. A, in HeLa cells GFP-Cx43 was co-localized with Cx43-mRFP at intracellular compartments and at plaque-like structures at cell-cell interfaces. B, N2A cell pairs expressing Cx43-mRFP (red) and GFP-Cx43 (green) co-localized (yellow) at cell-cell interfaces were chosen for patch clamp recordings. The phase image denotes the cellular architecture and location of the cell-cell interface. Bar = 10 µm. C, gap junctional conductance (Gj) was normalized to control condition (expressing Cx43-mRFP alone, the averaged intercellular coupling conductance for these cell pairs was measured to be 76 nS) and presented as a mean ± S.E., n = 22 for each data point. *, p < 0.05; ***, p < 0.001. Expression of Cx43-mRFP or Cx43-GFP exhibited similar macroscopic junctional conductance (white bar and hatched bar, respectively). With the increase of GFP-Cx43 amount, Cx43-mRFP-based gap junctional conductance was decreased in a dose-dependent manner. D, dose-inhibition relationship was plotted and fitted to a single exponential curve, which yielded an estimated ID50 of 1.97 (GFP-Cx43:Cx43-mRFP ratio).

 
With these findings in hand we now proceeded to address the mechanism of the specific dominant negative action of connexin variants under conditions where the dosage of each variant could be controlled. To address these issues, we developed a two color system where non-functional GFP-Cx43 and functional Cx43-mRFP (monomeric red fluorescent protein) (32) could be optically separated, quantitatively co-localized, and assessed for the level of electrical gap junction coupling. First, we co-transfected HeLa cells with Cx43-mRFP and GFP-Cx43 using various cDNA ratios (0.6-15 µg per transfection). Co-localization of GFP-Cx43 and Cx43-mRFP was often observed in the plaque-like structures at the cell-cell interfaces and in intracellular organelles in HeLa cells (Fig. 4A). Importantly, we observed that the amount of GFP-Cx43 and Cx43-mRFP expression, as measured by the respective fluorescence intensities in the putative GJ plaque regions, was approximately correlated with the amount of cDNA used in transfection. Similar observations were also obtained in N2A cells, which are more readily amenable to patch clamp recordings (Fig. 4B). On its own, Cx43-mRFP produced a robust macroscopic gap junctional coupling conductance nearly identical to that obtained from the recordings in Cx43-GFP-expressing N2A cells (Fig. 4C). In the co-transfection experiments, the Cx43-mRFP cDNA was kept constant at 3 µg, and various amounts of GFP-Cx43 cDNA were used to assess the relative amounts of GFP-Cx43 needed to inhibit the function of Cx43-mRFP. Our studies demonstrated that the cDNA dose ratio of GFP-Cx43 to Cx43-mRFP that revealed a significant reduction in coupling conductance was 1:1. A maximum of 84% reduction was seen when the cDNA dose ratio of GFP-Cx43 to Cx43-mRFP reached 4:1 (Fig. 4C). Based on these data, a dose-inhibition relationship was plotted and fitted to an exponential curve with an estimated ID50 of 1.97. The estimated ID50 is probably an overestimation, because we observed a decrease in the red fluorescence intensity level of Cx43-mRFP in the putative plaque area of HeLa cells when the dose of GFP-Cx43 cDNA was increased (data not shown). Consequently, with this proof of principle finding in hand we were in a position to examine the potency of naturally occurring Cx43 mutants found in ODDD patients.


Figure 5
View larger version (18K):
[in this window]
[in a new window]

  		FIGURE 5.
The ODDD-linked mutants, G21R and G138R, exerted a selective dominant negative effect on endogenous Cx43-based GJIC. When expressed in BICR-M1RK cells, the G21R and G138R mutants significantly inhibited GJIC mediated by endogenous Cx43 (A). In contrast, the expression of the mutants did not block GJIC mediated via Cx32 in HeLa-Cx32D cells (B). Data were normalized to control and presented as mean ± S.E., the averaged intercellular coupling conductance for pairs of BICR-M1Rk and HeLa-Cx32D was 80 and 43 nS, respectively, 10 ≤ n ≤ 16 for each data point. *, p < 0.05; ***, p < 0.001.

 


Figure 6
View larger version (119K):
[in this window]
[in a new window]

  		FIGURE 6.
G21R and G138R mutants were co-localized with wild-type Cx43 at cell-cell interfaces and intracellular compartments. When G21R-GFP or G138R-GFP was co-transfected with Cx43-mRFP at a 1 to 1 cDNA ratio into GJIC-deficient HeLa cells, the mutant and wild-type Cx43 exhibited similar distribution patterns at the cell surface and intracellular compartments (A-D, open arrowheads). Intermixing (open arrows) of mutant and wild-type Cx43 was prevalent in the GJ plaque regions (B and D). Bar = 5 µm.

 
ODDD-linked Human Cx43 Mutants, G21R and G138R, Exerted a Dominant Negative Effect on Endogenous Cx43 but Not Cx32—In our previous study we demonstrated that the human Cx43 ODDD mutants, G21R and G138R, reduced the ability of NRK cells to pass Lucifer yellow to adjacent cells (18). To determine if this dominant negative effect of G21R and G138R extended to electrical coupling and was connexin-selective, BICR-M1Rk and HeLa-Cx32D cells were transfected with cDNAs encoding G21R-GFP or G138R-GFP, and dual whole cell patch clamp recordings were acquired. Cx43-positive BICR-M1Rk cells expressing either the G21R-GFP or G138R-GFP mutant exhibited a marked decrease in gap junctional conductance (Fig. 5A). However, neither G21R nor G138R reduced the intercellular coupling between Cx32-expressing HeLa-Cx32D cells (Fig. 5B). These results suggest that both G21R and G138R mutants display dominant negative action only when co-expressed with endogenous Cx43 but not with Cx32.

G21R and G138R Mutants Co-localize with Wild-type Cx43 at Cell-Cell Interfaces and in Intracellular Compartments—When G21R-GFP or G138R-GFP was co-expressed with Cx43-mRFP in GJIC-deficient HeLa (Fig. 6) or N2A (data not shown) cells, both mutants co-localized with wild-type Cx43 within different intracellular compartments (open arrowheads) and at cell-cell interfaces to form plaque-like structures (open arrows). These results suggest that the G21R or G138R mutants shared common secretory and internalization pathways as their wild-type Cx43 counterpart (Fig. 6, A and C). High frequencies of G21R or G138R mutant co-localized with wild-type Cx43 within the gap junction plaques and other intracellular compartments (Fig. 6, B and D) provide indirect evidence that both mutants and wild-type Cx43 species have the potential and opportunity to co-interact to form heteromeric hemichannels and mixed gap junction channels.

The Incorporation of G21R and G138R Mutants into Putative Gap Junction Plaques with Wild-type Cx43 Was Dose-dependent—To quantitatively study the co-distribution of mutant and wild-type Cx43, cDNAs encoding GFP-tagged mutant and mRFP-tagged wild-type Cx43 were mixed in various ratios and transfected into HeLa cells. These co-transfection experiments provided an ideal system for monitoring the expression and interaction between mutant and wild-type Cx43 (Fig. 7A and 8A). Although similar co-distribution patterns between the mutant and wild-type Cx43 were observed within some intracellular compartments and at cell-cell interfaces, the quantitative analysis was only carried out in putative gap junction plaque regions.


Figure 7
View larger version (45K):
[in this window]
[in a new window]

  		FIGURE 7.
G21R was incorporated into gap junctional plaques with wild-type Cx43 in a dose-dependent manner. When G21R-GFP was co-transfected with Cx43-mRFP into HeLa cells, the amount of G21R-GFP cDNA used for transfection was proportional to the average total green fluorescence detected in the putative GJ plaques (A). The red lines in the images indicate the profiles where fluorescence intensity was measured, and the red and green traces depict fluorescence intensity profiles for Cx43-mRFP and G21R-GFP, respectively (A). To accurately measure the red and green fluorescence intensities within putative GJ plaques, ROIs were carefully selected and the mean fluorescence intensities were then computed (B). The averaged fluorescence intensity ratios of G21R-GFP over Cx43-mRFP were calculated and plotted (C). Each bar represents the average of 20 GJ plaques analyzed for each experimental condition. Bar = 5 µm.

 


Figure 8
View larger version (39K):
[in this window]
[in a new window]

  		FIGURE 8.
G138R was incorporated into gap junctional plaques with wild-type Cx43 in a dose-dependent manner. G138R-GFP showed a dose dependence in mixing with Cx43-mRFP in putative GJ plaques. A, the red lines in the image panels indicate the profiles used for measuring fluorescence intensities, and the red and green traces below the images represent the fluorescence intensities for Cx43-mRFP and G21R-GFP, respectively. B, to precisely quantify the red and green fluorescence intensities at the putative gap junction plaques, ROIs were selected and the mean fluorescence intensities were computed. C, the averaged fluorescence intensity ratios of G138R-GFP to Cx43-mRFP were calculated and plotted. Each bar depicts the average of 20 GJ plaques selected under each experimental condition. Bar = 5 µm.

 
In cells expressing Cx43-mRFP only, the fluorescence intensity profile revealed a red fluorescence peak at the cell-cell interface indicative of a gap junction plaque region. While maintaining the amount of Cx43-mRFP cDNA constant (3 µg), we gradually increased the amount of GFP-tagged mutant cDNAs used in co-transfection resulting in a proportional increase in the intermixing of both red- and green-tagged Cx43 variants at the putative gap junction plaques (Figs. 7A and 8A). For accurate quantification, we carefully selected putative GJ plaque areas from the co-expressing cells as regions of interest (ROIs) (Figs. 7B and 8B) and calculated the fluorescence intensity ratios (green channel/red channel) (Figs. 7C and 8C). A notable but relatively equal decrease in the red fluorescence intensity level of Cx43-mRFP in the ROIs was observed when the dose of G21R-GFP or G138R-GFP cDNA was increased. Interestingly, when the same amounts of cDNAs were used for the mutants and wild-type Cx43, the mixture of both variants within a plaque were equal, suggesting that the mutants were equally efficient at being assembled into a gap junction plaque as wild-type Cx43. These data further support that the G21R-GFP and G138R-GFP mutants exhibit no detectable trafficking or assembly defects. Similar qualitative results were observed when N2A cells were used.

G21R, G138R, and GFP-Cx43 Physically Interact with Wild-type Cx43—We have provided morphological evidence that G21R-GFP, G138R-GFP, and GFP-Cx43 were able to co-localize with either endogenous Cx43 or Cx43-mRFP in intracellular compartments and within the gap junction plaque-like structures. However, to assess the evidence for a molecular interaction we performed co-immunoprecipitation experiments. G21R-GFP, G138R-GFP, GFP-Cx43, or Cx32-GFP was co-transfected with Cx43-mRFP in HeLa cells and lysates from these cells were used to immunoprecipitate the various GFP-tagged proteins. As shown in Fig. 9B, Cx32-GFP and all the GFP-tagged Cx43 constructs were efficiently immunoprecipitated using a specific anti-GFP antibody. GFP-Cx43 was detected at a higher relative molecular weight than the other Cx43-GFP tagged mutants due to an increase in the length of the linker sequence between GFP and Cx43 (see Fig. 1A). Importantly, Cx43-mRFP was detected in the GFP immunoprecipitates from the cells overexpressing wild-type Cx43-GFP, as well as the dominant negative Cx43 mutants G21R-GFP, G138R-GFP, and GFP-Cx43 (Fig. 9A). However, it was not co-immunoprecipitated with Cx32-GFP, even though there was evidence of Cx43-mRFP in the cell lysate (Fig. 9C). The fact that Cx43-mRFP co-immunoprecipitated with wild-type and Cx43 mutants suggested that ODDD-linked Cx43 mutants physically interact with wild-type Cx43.


Figure 9
View larger version (39K):
[in this window]
[in a new window]

  		FIGURE 9.
Wild-type and Cx43 mutants co-immunoprecipitate. Lysates from HeLa cells overexpressing Cx43-mRFP together with Cx32-GFP or Cx43 variants (Cx43-GFP, G21R-GFP, G138R-GFP, and GFP-Cx43) were subjected to immunoprecipitation (IP) using a monoclonal antibody specific for GFP. The presence of Cx43-mRFP in the immunoprecipitates was monitored by immunoblotting (IB). A, Cx43-mRFP was strongly detected in the GFP immunoprecipitates from cell overexpressing Cx43-GFP, as well as the Cx43 dominant negative mutants. As opposed to the cells overexpressing the GFP-tagged Cx43 variants, Cx43-mRFP was not detected in the GFP immunoprecipitates from HeLa overexpressing Cx32-GFP. The two lower bands detected using mRFP antibody in this experiment are likely nonspecific, because their molecular weights are less than Cx43-mRFP. In addition, these bands were not detected in the cell lysates using mRFP antibodies. B, immunoblotting for GFP revealed the presence of Cx32-GFP and the GFP-tagged variants of Cx43 at all the correct molecular weights, including the larger of GFP-Cx43 due to lengthy linker sequences used in the fusion protein. The presence of the IgG band is also noted. C, as revealed by Western blotting, Cx43-mRFP was expressed in similar levels in the cell lysates used for the immunoprecipitations.

 
G21R and G138R Mutants Exhibited Differential Potencies in Inhibiting Cx43-based GJIC—Our two-color labeling strategy in combination with dual whole cell patch clamp was used to investigate the various molar ratio of mutant to wild-type Cx43 required to inhibit Cx43 gap junction function. Cx43-mRFP cDNA (3 µg) was mixed with various amounts of G21R-GFP or G138R-GFP cDNA to achieve various ratios of mutant to wild-type Cx43 expression. The mixed cDNAs were then co-transfected into GJIC-deficient N2A cells, and individual cell pairs with clear co-localized red and green gap junction plaques at the cell-cell interfaces were selected for recording. The G21R mutant exerted a strong inhibitory action over the ability of Cx43-mRFP to form functional gap junction channels (Fig. 10A). At a G21R mutant to Cx43-mRFP cDNA ratio of 1:5, 63% of junctional conductance (Gj) was blocked. At a ratio of 1:1, over 90% of total Gj was blocked (Fig. 10A). However, in the case of the G138R mutant, a higher concentration of the mutant was required to achieve partial or nearly complete inhibition of Cx43-mRFP function (Fig. 10C). Upon constructing exponential dose-inhibition curves, our data revealed that the G21R mutant had an estimated ID50 of 0.126 (Fig. 10B), whereas the ID50 for the G138R mutant was 0.393 (Fig. 10D), indicating that these mutants exhibited differential potencies. We caution that the estimated ID50 is a measurement of relative potency between mutants and may not reflect the true ID50 of each mutant.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
The discovery of disease linkages among germ line mutations in genes that encode the connexin family of gap junction proteins has raised considerable interest into the consequences of such mutations on connexin trafficking and function. Importantly, we now know that multiple single missense mutations in GJA1 can lead to partial or complete loss of Cx43 channel function manifesting as an array of ODDD patients with diverse clinical symptoms. Although ODDD-linked mutant overexpression systems have provided several insights into the consequences of a specific mutation on channel function, it is more difficult to extrapolate these findings to the clinical setting, because the mutant to wild-type Cx43 ratios do not mimic the expected human condition. These limitations are further exacerbated by the fact that no mouse model of ODDD has been reported involving a known ODDD-linked human Cx43 mutation. In the present study we developed a novel approach to circumvent these limitations by combining a fluorescence-based analysis of mutant to wild-type Cx43 ratios with highly sensitive dual whole cell patch clamp to assess coupling conductance. Upon establishing proof-of-principle analysis using functional and non-functional GFP-connexin chimeras we used this approach to demonstrate the variable dominant negative potencies of two distinct ODDD-linked mutants.

Fluorescent Protein-tagged Connexin Analysis Reveals Selective Molecular Mechanisms of Dominant Negative Inhibition of Gap Junction Channels—Various studies from our laboratory and others have demonstrated that connexins with GFP tagged to the carboxyl terminus (Cxs-GFP) largely mimic the biosynthesis, trafficking, assembly, and function of their wild-type counterparts (24-30). However, in an earlier study, we showed that fusion of GFP to the amino terminus of Cx43 inhibited the ability of this fusion protein to allow dye coupling with Lucifer yellow even though it could successfully reach the plasma membrane and cluster into putative gap junction plaques (29). Not surprisingly, a subsequent study using patch clamp analysis revealed that GFP-Cx43 could form neither functional hemichannels nor whole channels (30). In our current study, we found that, when GFP was fused to a second connexin, Cx32, channel function was also abolished. Given that Cx32 belongs to the beta-subgroup of the connexin family, our findings suggest that GFP tagging to the amino terminus of possibly any connexin may inhibit its function. The mechanism of this inhibition of connexin function is not known but may be related to steric hindrance associated with the bulky nature of GFP when tagged to the amino terminus. The resulting gap junction plaque-like structures that do form may contain aberrant channels where the pore is permanently blocked by GFP. However, when GFP is tagged to the carboxyl terminus of connexins, it is predicted that there is sufficient space available to arrange six GFP molecules (37) in a configuration where the pore remains open (26) even in cases where GFP is tagged to the very short carboxyl tail of Cx26 (26, 27, 29).


Figure 10
View larger version (17K):
[in this window]
[in a new window]

  		FIGURE 10.
The ODDD-linked Cx43 mutants, G21R and G138R, exhibited different potencies in inhibiting Cx43-mediated GJIC. GJIC-deficient N2A cells were transfected with mixed cDNAs encoding G21R-GFP or G138R-GFP and Cx43-mRFP in various ratios. A constant amount (3 µg) of Cx43-mRFP cDNA was used in all experiments together with various amounts of G21R-GFP or G138R-GFP cDNAs ranging from 0 to 15 µg. Isolated cell pairs with co-localized green and red plaques at the cell-cell interfaces were selected for dual whole cell patch clamp recording. Gap junctional conductance was determined for G21R (A) and G138R (C) transfected cell pairs. Data were normalized to control and presented as mean ± S.E., n = 20 for each data point. *, p < 0.05; **, p < 0.01; and ***, p < 0.001. The averaged junctional conductance for the control pairs in panel A was 43 nS and in panel C was 55 nS. Dose-inhibition relationships were fitted to single exponential curves for G21R (B) and G138R (D). ID50, dose ratio of the mutant Cx43 cDNA to wild-type Cx43 cDNA.

 
GFP-Cx43 dose-dependently inhibited Cx43-based, but not Cx32-based GJIC. Possibly the simplest interpretation for these findings would be that both non-functional and functional fluorescent protein-tagged Cx43 fusion proteins compete for limited cell-cell surface-interacting domains where gap junctions can be assembled (i.e. competitive mechanism). In the event that non-functional Cx43 variants occupy these cell-cell-interacting areas, this could in principle inhibit the function of wild-type Cx43.

Although the competitive mechanism for how non-functional GFP-Cx43 may inhibit GJIC is plausible, this mechanism is not supported by the fact that GFP-Cx32 could not inhibit Cx43-based GJIC, or vice versa. This is consistent with previous findings where Cx43 was determined not to co-oligomerize with Cx32 (38, 39). Consequently, we propose that the mechanism of inhibition is based on co-oligomerization of non-functional connexin chimeras with their compatible functional connexin counterparts. Our co-immunoprecipitation experiments demonstrated that GFP-Cx43 was able to physically interact with wild-type Cx43. Mixed oligomers of functional and non-functional connexins may sufficiently alter the overall structure of the connexon such that connexons from apposing cells cannot properly dock or unable to form a conducting pore in a connexon.

The proposed mixed oligomer mechanism could also explain the dose dependence of GFP-Cx43, because the dominant negative potency on Cx43 was found to increase with the addition of more GFP-Cx43 subunits. According to probability theory, and based on the co-expression of two connexin variants within the same cell and hexameric arrangement of connexins into connexons, small changes in the expression ratio of connexin variants could lead to substantial changes in connexon composition (8). By manipulating the relative expression levels of functional and non-functional connexins, we were able to estimate the potency of dominant negative activities on gap junction channels. In addition, our dual-color co-expression system rendered visualization of the expression of both connexins possible, where cell imaging and patch clamp recording were performed in parallel in two GJIC-incompetent cell lines.

ODDD-linked Cx43 Mutants Have Distinct Potencies in Inhibiting Wild-type Cx43—To date, over 30 Cx43 mutations have been identified as ODDD-linked, and 12 of them have been at least partially characterized (16-20). In theory, mutations in the GJA1 gene encoding Cx43 that cause ODDD may affect one or several steps in the Cx43 life cycle that ultimately result in three phenotypic channel outcomes that include: gain of function, impaired function, or loss of function. Among loss-of-function mutants, some may exhibit defective transport and assembly properties and the potential to be rescued by co-expressed wild-type connexins, whereas others may properly traffic in the cell but form functionally inactive gap junction channels. It is also possible that some of the loss-of-function mutants will have dominant negative effects on wild-type connexin(s) via co-oligomerization or sequestration. In vitro studies, including ours, found that several ODDD Cx43 mutants, such as Y17S, G21R, A40V, Q49K, L90V, I130T, K134E, and G138R, retain the capacity to traffic to the cell surface and form gap junction plaque-like structures, but none of them form functional channels (16-19). Some mutants, F52dup, R202H, and V216L, exhibit reduced ability to form gap junction plaques (17, 19). Intriguingly, F52dup and R202H could be rescued to cell-cell junctions and restore functional coupling when co-expressed with wild-type Cx43 (19). On the other hand, we have reported that a frameshift mutant (fs260) not only exhibits impaired trafficking, but also retards wild-type Cx43 assembly into gap junction plaques (20). In other studies, a mutation in the mouse gene encoding Cx43 (G60S) resulted in a mutant mouse line that mimicked many of the phenotypic symptoms of human ODDD (40). In these mice there was an ~80% reduction in Cx43 protein expression in both the heart and ovaries, which is most likely a result of impaired biosynthesis and/or a drastically accelerated degradation process (40).

In the current study, we found that both G21R and G138R mutants demonstrated a selective dominant negative effect on wild-type Cx43, but not Cx32. Both mutants appeared to share similar trafficking pathways as Cx43 forming gap junctional plaques in a dose-dependent manner. Collectively, our results are entirely consistent with the ODDD-linked Cx43 mutants (G21R and G138R) exerting their selective dominant negative effect by randomly co-oligomerizing with wild-type Cx43, which was supported by co-immunoprecipitation studies.

G21R and G138R are both missense mutations where a neutral glycine residue is replaced by a positively charged arginine residue. Without a side chain, glycine can rotate easily and often exists in the turn regions between {alpha} helices (41). Arginine is highly hydrophilic and commonly seen at the aqueous surface of proteins, unless it's positively charged side chain (7 ~ 8Åin length) is sequestered by strong negative charges from protein interiors (42). Consequently, the major changes in these amino acid properties are predicted to alter the characteristics of the M1 and intracellular loop domains of Cx43. Because M1 is thought to contribute to the lining of the gap junction pore, and the G21R mutant is very close to the pore-opening cytoplasmic face, it is reasonable to predict that the long side chain of arginine might partially or completely block the channel opening. Alternatively, the arginine could greatly alter the conformation of the gap junction pore resulting in a narrower or closed channel.

In the case of the G138R mutant, the amino acid substitution occurs within the second half of the intracellular loop of Cx43. An early report revealed that the interaction between the intracellular loop and the carboxyl terminus tail is important to gap junction function, because swapping these two domains between Cx32 and Cx43 yielded functionally impaired chimeric Cx32/Cx43 (43). Likewise, a subsequent study demonstrated that a deletion of seven amino acids from this region resulted in a dominant negative mutation (Cx43{Delta}130-136) (44). This deletion mutant could be transported to cell surface but showed abolished permeability to Lucifer yellow. According to the "ball-and-chain" model of pH-gated channel closure (45), the carboxyl terminus of Cx43 acts as a gating particle upon acidification and two intramolecular regions in the intracellular loop (119-144) serve as the receptor (45, 46). Thus the G138R mutation might drastically alter the electrostatic microenvironment in the intracellular loop region and severely interfere with these delicate interactions. Likewise, it was not surprising to see that the G138R mutant was incapable of forming functional homomeric channels, and dose-dependently inhibited Cx43 function when heteromeric oligomers were predicted to form.

Many connexin mutants linked to human diseases exhibit both dominant and possibly trans-dominant negative effects in vitro or in vivo (36, 47-51). One study reported that only one P88S Cx50 mutant subunit, oligomerized with wild-type Cx50, was sufficient to abolish Cx50 channel function (52). In our current study, we assessed the effect of ODDD-linked mutants on the total levels of GJIC using a unique two-color fluorescent and dual patch clamp approach. We first established that both mutants and wild-type Cx43 have equal opportunity to intermix and ultimately co-exist in the same putative gap junction plaques at ratios roughly similar from the amounts of cDNA used in the transfection assay. G21R and G138R exhibited similar dose dependence, indicating that wild-type Cx43 has equal apparent "affinity" or "ability" to intermix with either mutant. Thus, the differential potency of each mutant reflects the ability of each mutant to impair the GJ channel and not on any differential ability to co-exist within a gap junction plaque. Regardless of whether the individual channels are partly or completely closed, these findings provide novel insights into dominant negative mechanism underlying ODDD. Our dosage experiments clearly demonstrate that the same amino acid residue substitutions in different Cx43 domains result in dominant negative mutants with distinct potencies. Thus, G21R is about 2 times more potent in inhibiting wild-type Cx43 function than G138R. We propose that the differential clinical manifestations of ODDD in patients harboring the G21R or G138R mutants may, in part, be related to the nature and position of the missense mutations. The correlation between the different potencies and the phenotypic variability presented in ODDD patients remains to be tested when more clinical information is available from a broader spectrum of patients.

Finally, it is important to note that ODDD patients have only one allele mutated and thus are expected to transcribe equal amounts of both mutant and wild-type Cx43 mRNAs. In the event that both transcripts are equally stable, the resulting mutant and wild-type Cx43 proteins translated would also be expected to be at the same level. Under these conditions, our results demonstrated that at a predicted 1:1 mutant to wild-type Cx43 ratio GJIC was reduced by 95 and 84% for the G21R and G138R mutants, respectively. Remarkably, our study would predict that ODDD patients are able to survive with only 5-16% normal Cx43-based GJIC essentially consistent with our previous findings in a mouse model of ODDD (40). Such patient survival may point to mechanisms where other members of the connexin family or other molecules (e.g. pannexins) are able to compensate for this drastic reduction in Cx43-based GJIC. Alternatively, under normal conditions, Cx43 may be produced in far excess of biological need. Moreover, it will be important to know if these same ODDD-linked Cx43 mutants have dominant negative effects on other compatible connexin subtypes. These questions remain to be elucidated in future studies.


   FOOTNOTES
 
* This work was supported by the Canadian Institutes of Health Research (CIHR) and Canada Research Chair program (to D. B. and D. W. L.) and a CIHR Strategic Training Scholarship (to S. L.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

2 Both authors contributed equally to this work. Back

1 To whom correspondence should be addressed: Dept. of Physiology and Pharmacology, University of Western Ontario, London, Ontario N6A 5C1, Canada. Tel.: 519-850-2569; Fax: 519-850-2562; E-mail: donglin.bai{at}schulich.uwo.ca.

3 The abbreviations used are: GJ, gap junction; GJIC, gap junctional intercellular communication; ODDD, oculodentodigital dysplasia; mRFP, monomeric red fluorescent protein; GFP, green fluorescent protein; NRK, normal rat kidney; ROIs, regions of interest. Back


   ACKNOWLEDGMENTS
 
We thank Paulina Fistouris for assistance in the microinjection experiments.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Kumar, N. M., and Gilula, N. B. (1996) Cell 84, 381-388[CrossRef][Medline] [Order article via Infotrieve]
  2. Saez, J. C., Berthoud, V. M., Branes, M. C., Martinez, A. D., and Beyer, E. C. (2003) Physiol. Rev. 83, 1359-1400[Abstract/Free Full Text]
  3. Laird, D. W., and Revel, J. P. (1990) J. Cell Sci. 97, 109-117[Abstract/Free Full Text]
  4. Falk, M. M., Buehler, L. K., Kumar, N. M., and Gilula, N. B. (1997) EMBO J. 16, 2703-2716[CrossRef][Medline] [Order article via Infotrieve]
  5. Willecke, K., Eiberger, J., Degen, J., Eckardt, D., Romualdi, A., Guldenagel, M., Deutsch, U., and Sohl, G. (2002) Biol. Chem. 383, 725-737[CrossRef][Medline] [Order article via Infotrieve]
  6. Sosinsky, G. E., and Nicholson, B. J. (2005) Biochim. Biophys. Acta 1711, 99-125[Medline] [Order article via Infotrieve]
  7. Evans, W. H., and Martin, P. E. (2002) Mol. Membr. Biol. 19, 121-136[CrossRef][Medline] [Order article via Infotrieve]
  8. Cottrell, G. T., and Burt, J. M. (2005) Biochim. Biophys. Acta 1711, 126-141[Medline] [Order article via Infotrieve]
  9. Herskowitz, I. (1987) Nature 329, 219-222[CrossRef][Medline] [Order article via Infotrieve]
  10. Sheppard, D. (1994) Am. J. Respir. Cell Mol. Biol. 11, 1-6[Medline] [Order article via Infotrieve]
  11. Kelsell, D. P., Di, W. L., and Houseman, M. J. (2001) Am. J. Hum. Genet. 68, 559-568[CrossRef][Medline] [Order article via Infotrieve]
  12. Chang, E. H., Van Camp, G., and Smith, R. J. (2003) Ear. Hear. 24, 314-323[CrossRef][Medline] [Order article via Infotrieve]
  13. Richard, G. (2003) Clin. Exp. Dermatol. 28, 397-409[CrossRef][Medline] [Order article via Infotrieve]
  14. Paznekas, W. A., Boyadjiev, S. A., Shapiro, R. E., Daniels, O., Wollnik, B., Keegan, C. E., Innis, J. W., Dinulos, M. B., Christian, C., Hannibal, M. C., and Jabs, E. W. (2003) Am. J. Hum. Genet. 72, 408-418[CrossRef][Medline] [Order article via Infotrieve]
  15. Frasson, M., Calixto, N., Cronemberger, S., de Aguiar, R. A., Leao, L. L., and de Aguiar, M. J. (2004) Ophthalmic Genet. 25, 227-236[CrossRef][Medline] [Order article via Infotrieve]
  16. Seki, A., Coombs, W., Taffet, S. M., and Delmar, M. (2004) Heart Rhythm 1, 227-233[CrossRef][Medline] [Order article via Infotrieve]
  17. McLachlan, E., Manias, J. L., Gong, X. Q., Lounsbury, C. S., Shao, Q., Bernier, S. M., Bai, D., and Laird, D. W. (2005) Cell Commun. Adhes. 12, 279-292[CrossRef][Medline] [Order article via Infotrieve]
  18. Roscoe, W., Veitch, G. I., Gong, X. Q., Pellegrino, E., Bai, D., McLachlan, E., Shao, Q., Kidder, G. M., and Laird, D. W. (2005) J. Biol. Chem. 280, 11458-11466[Abstract/Free Full Text]
  19. Shibayama, J., Paznekas, W., Seki, A., Taffet, S., Jabs, E. W., Delmar, M., and Musa, H. (2005) Circ. Res. 96, e83-91[Abstract/Free Full Text]
  20. Gong, X. Q., Shao, Q., Lounsbury, C. S., Bai, D., and Laird, D. W. (2006) J. Biol. Chem. 281, 31801-31811[Abstract/Free Full Text]
  21. Lai, A., Le, D. N., Paznekas, W. A., Gifford, W. D., Jabs, E. W., and Charles, A. C. (2006) J. Cell Sci. 119, 532-541[Abstract/Free Full Text]
  22. Boyadjiev, S. A., Jabs, E. W., LaBuda, M., Jamal, J. E., Torbergsen, T., Ptacek, L. J., 2nd, Rogers, R. C., Nyberg-Hansen, R., Opjordsmoen, S., Zeller, C. B., Stine, O. C., Stalker, H. J., Zori, R. T., and Shapiro, R. E. (1999) Genomics 58, 34-40[CrossRef][Medline] [Order article via Infotrieve]
  23. Shapiro, R. E., Griffin, J. W., and Stine, O. C. (1997) Am J. Med. Genet. 71, 36-41[CrossRef][Medline] [Order article via Infotrieve]
  24. Jordan, K., Solan, J. L., Dominguez, M., Sia, M., Hand, A., Lampe, P. D., and Laird, D. W. (1999) Mol. Biol. Cell 10, 2033-2050[Abstract/Free Full Text]
  25. Bukauskas, F. F., Jordan, K., Bukauskiene, A., Bennett, M. V., Lampe, P. D., Laird, D. W., and Verselis, V. K. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, 2556-2561[Abstract/Free Full Text]
  26. Falk, M. M. (2000) J. Cell Sci. 113, 4109-4120[Abstract]
  27. Paemeleire, K., Martin, P. E., Coleman, S. L., Fogarty, K. E., Carrington, W. A., Leybaert, L., Tuft, R. A., Evans, W. H., and Sanderson, M. J. (2000) Mol. Biol. Cell 11, 1815-1827[Abstract/Free Full Text]
  28. Bukauskas, F. F., Bukauskiene, A., Bennett, M. V., and Verselis, V. K. (2001) Biophys. J. 81, 137-152[Medline] [Order article via Infotrieve]
  29. Laird, D. W., Jordan, K., Thomas, T., Qin, H., Fistouris, P., and Shao, Q. (2001) Microsc. Res. Tech 52, 263-272[CrossRef][Medline] [Order article via Infotrieve]
  30. Contreras, J. E., Saez, J. C., Bukauskas, F. F., and Bennett, M. V. (2003) Proc. Natl. Acad. Sci. U. S. A. 100, 11388-11393[Abstract/Free Full Text]
  31. Thomas, T., Jordan, K., Simek, J., Shao, Q., Jedeszko, C., Walton, P., and Laird, D. W. (2005) J. Cell Sci. 118, 4451-4462[Abstract/Free Full Text]
  32. Campbell, R. E., Tour, O., Palmer, A. E., Steinbach, P. A., Baird, G. S., Zacharias, D. A., and Tsien, R. Y. (2002) Proc. Natl. Acad. Sci. U. S. A. 99, 7877-7882[Abstract/Free Full Text]
  33. Laird, D. W., Castillo, M., and Kasprzak, L. (1995) J. Cell Biol. 131, 1193-1203[Abstract/Free Full Text]
  34. Fujikura, Y., Ohta, H., Hirai, T., and Fukumoto, T. (1993) Anat. Rec. 235, 335-341[CrossRef][Medline] [Order article via Infotrieve]
  35. Musa, H., Fenn, E., Crye, M., Gemel, J., Beyer, E. C., and Veenstra, R. D. (2004) J. Physiol. 557, 863-878[Abstract/Free Full Text]
  36. Gollob, M. H., Jones, D. L., Krahn, A. D., Danis, L., Gong, X. Q., Shao, Q., Liu, X., Veinot, J. P., Tang, A. S., Stewart, A. F., Tesson, F., Klein, G. J., Yee, R., Skanes, A. C., Guiraudon, G. M., Ebihara, L., and Bai, D. (2006) N. Engl. J. Med. 354, 2677-2688[Abstract/Free Full Text]
  37. Ormo, M., Cubitt, A. B., Kallio, K., Gross, L. A., Tsien, R. Y., and Remington, S. J. (1996) Science 273, 1392-1395[Abstract]
  38. Elfgang, C., Eckert, R., Lichtenberg-Frate, H., Butterweck, A., Traub, O., Klein, R. A., Hulser, D. F., and Willecke, K. (1995) J. Cell Biol. 129, 805-817[Abstract/Free Full Text]
  39. White, T. W., Paul, D. L., Goodenough, D. A., and Bruzzone, R. (1995) Mol. Biol. Cell 6, 459-470[Abstract]
  40. Flenniken, A. M., Osborne, L. R., Anderson, N., Ciliberti, N., Fleming, C., Gittens, J. E., Gong, X. Q., Kelsey, L. B., Lounsbury, C., Moreno, L., Nieman, B. J., Peterson, K., Qu, D., Roscoe, W., Shao, Q., Tong, D., Veitch, G. I., Voronina, I., Vukobradovic, I., Wood, G. A., Zhu, Y., Zirngibl, R. A., Aubin, J. E., Bai, D., Bruneau, B. G., Grynpas, M., Henderson, J. E., Henkelman, R. M., McKerlie, C., Sled, J. G., Stanford, W. L., Laird, D. W., Kidder, G. M., Adamson, S. L., and Rossant, J. (2005) Development 132, 4375-4386[Abstract/Free Full Text]
  41. Liew, A. W.-C., Yan, H., and Yang, M. (2005) in Bioinformatics Technologies (Chen, Y.-P. P., ed) pp. 63-116, Springer-Verlag Ltd., Berlin
  42. Engelman, D. M., Steitz, T. A., and Goldman, A. (1986) Annu. Rev. Biophys. Biophys. Chem. 15, 321-353[CrossRef][Medline] [Order article via Infotrieve]
  43. Bruzzone, R., White, T. W., and Paul, D. L. (1994) J. Cell Sci. 107, 955-967[Abstract]
  44. Krutovskikh, V. A., Yamasaki, H., Tsuda, H., and Asamoto, M. (1998) Mol. Carcinog. 23, 254-261[CrossRef][Medline] [Order article via Infotrieve]
  45. Morley, G. E., Taffet, S. M., and Delmar, M. (1996) Biophys. J. 70, 1294-1302[Medline] [Order article via Infotrieve]
  46. Duffy, H. S., Sorgen, P. L., Girvin, M. E., O'Donnell, P., Coombs, W., Taffet, S. M., Delmar, M., and Spray, D. C. (2002) J. Biol. Chem. 277, 36706-36714[Abstract/Free Full Text]
  47. Rouan, F., White, T. W., Brown, N., Taylor, A. M., Lucke, T. W., Paul, D. L., Munro, C. S., Uitto, J., Hodgins, M. B., and Richard, G. (2001) J. Cell Sci. 114, 2105-2113[Abstract/Free Full Text]
  48. Kudo, T., Kure, S., Ikeda, K., Xia, A. P., Katori, Y., Suzuki, M., Kojima, K., Ichinohe, A., Suzuki, Y., Aoki, Y., Kobayashi, T., and Matsubara, Y. (2003) Hum. Mol. Genet. 12, 995-1004[Abstract/Free Full Text]
  49. Marziano, N. K., Casalotti, S. O., Portelli, A. E., Becker, D. L., and Forge, A. (2003) Hum. Mol. Genet. 12, 805-812[Abstract/Free Full Text]
  50. Fiorini, C., Mograbi, B., Cronier, L., Bourget, I., Decrouy, X., Nebout, M., Ferrua, B., Malassine, A., Samson, M., Fenichel, P., Segretain, D., and Pointis, G. (2004) J. Cell Sci. 117, 4665-4672[Abstract/Free Full Text]
  51. Thomas, T., Telford, D., and Laird, D. W. (2004) J. Biol. Chem. 279, 19157-19168[Abstract/Free Full Text]
  52. Pal, J. D., Berthoud, V. M., Beyer, E. C., Mackay, D., Shiels, A., and Ebihara, L. (1999) Am. J. Physiol. 276, C1443-C1446[Medline] [Order article via Infotrieve]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 J. Cell Sci.Home page
J. Simek, J. Churko, Q. Shao, and D. W. Laird
Cx43 has distinct mobility within plasma-membrane domains, indicative of progressive formation of gap-junction plaques
J. Cell Sci., February 15, 2009; 122(4): 554 - 562.
[Abstract] [Full Text] [PDF]

Home page
 Cardiovasc ResHome page
J. L. Manias, I. Plante, X.-Q. Gong, Q. Shao, J. Churko, D. Bai, and D. W. Laird
Fate of connexin43 in cardiac tissue harbouring a disease-linked connexin43 mutant
Cardiovasc Res, December 1, 2008; 80(3): 385 - 395.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
D. W. Laird
Closing the Gap on Autosomal Dominant Connexin-26 and Connexin-43 Mutants Linked to Human Disease
J. Biol. Chem., February 8, 2008; 283(6): 2997 - 3001.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
S. Langlois, A. C. Maher, J. L. Manias, Q. Shao, G. M. Kidder, and D. W. Laird
Connexin Levels Regulate Keratinocyte Differentiation in the Epidermis
J. Biol. Chem., October 12, 2007; 282(41): 30171 - 30180.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
282/26/19190    most recent
M609653200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gong, X.-Q.
Right arrow Articles by Laird, D. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gong, X.-Q.
Right arrow Articles by Laird, D. W.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Molecular and Cellular Proteomics 
 Journal of Lipid Research   ASBMB Today 
Copyright © 2007 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement