点击显示 收起
The Departments of Pharmacology (J.S., A.S., M.D., H.M.)
Microbiology/Immunology (S.T.), S.U.N.Y. Upstate Medical University, Syracuse, NY 13210. Institute of Genetic Medicine, Department of Pediatrics (W.P., E.W.)
Medicine and Surgery (E.W.), Johns Hopkins University School of Medicine, Baltimore, MD.
Abstract
Specific mutations in GJA1, the gene encoding the gap junction protein connexin43 (Cx43), cause an autosomal dominant disorder called oculodentodigital dysplasia (ODDD). Here, we characterize the effects of 8 of these mutations on Cx43 function. Immunochemical studies have shown that most of the mutant proteins formed gap junction plaques at the sites of cell-cell apposition. However, 2 of the mutations (a codon duplication in the first extracellular loop, F52dup, and a missense mutation in the second extracellular loop, R202H, produced full-length connexins that failed to properly form gap junction plaques. Cx43 proteins containing ODDD mutations found in the N-terminus (Y17S), first transmembrane domain (G21R, A40V), second transmembrane domain (L90V), and cytoplasmic loop (I130T, K134E) do form gap junction plaques but show compromised channel function. L90V, I130T, and K134E demonstrated a significant decrease in junctional conductance relative to Cx43WT. Mutations Y17S, G21R, and A40V demonstrated a complete lack of functional electrical coupling even in the presence of significant plaque formation between paired cells. Heterologous channels formed by coexpression of Cx43WT and mutation R202H resulted in electrically functional gap junctions that were not permeable to Lucifer yellow. Therefore, the mutations found in ODDD not only cause phenotypic variability, but also result in various functional consequences. Overall, our data show an extensive range of molecular phenotypes, consistent with the pleiotropic nature of the clinical syndrome as a whole.
Key Words: connexin43 ODDD gap junctions
Introduction
Gap junctions are intercellular channels that provide a low resistance pathway coupling the cytoplasmic compartments of adjacent cells. They establish a pathway for cell-to-cell communication by allowing the direct exchange of ions, metabolites, and small molecules of up to 1kDa in size. Their role is vital for normal organ development1,2 and for coordinating and regulating specific functions in mature tissue.3,4,5 Each complete gap junction channel is composed of 2 hemichannels (connexons), 1 contributed by each cell of an adjacent pair that align in the extracellular space to create an intercellular hydrophilic pathway. Connexons are, in turn, made up of 6 intramembrane protein subunits termed connexins (Cx). Hydropathy plots of connexins predict a protein with 4 transmembrane domains (M1-M4), 2 extracellular loops (E1 and E2), 1 cytoplasmic loop (CL), and cytoplasmic amino and carboxyl termini (NT and CT). There are 21 different connexins identified in the human genome,6 each characterized by particular tissue distribution, channel conductance, molecular weight, transjunctional voltage dependence of gating, and modulation by physiological factors.7 Connexin isotypes are classified according to the predicted molecular weight of the gene product. Hence, Cx43 refers to a connexin molecule with a predicted molecular weight of 43 kDa. Cx43 is the most abundant connexin in the cardiovascular system. Transgenic animal models show that absence of Cx43 leads to cardiac malformations1 and significant slowing of conduction velocity.8 Yet, expression of Cx43 by 1 allele maintains normal action potential propagation and cardiogenesis.9 A number of cardiac diseases lead to remodeling of Cx43.10,11,12
Connexins play an important role in normal physiology and mutations in their primary sequence underlie a wide spectrum of human diseases. Mutations in connexin proteins have been implicated in disorders of the skin, hearing, vision, and nervous systems.13,14,15,16 Recently, an autosomal dominant pleiotropic disorder called oculodentodigital dysplasia (ODDD) has been linked to mutations in Cx43. This disorder presents wide intra- and inter-familial phenotypic variability and is associated with ocular, dental, digital, and craniofacial anomalies, neurodegeneration and cardiac abnormalities. Originally, 16 different missense mutations and 1 codon duplication were identified in 17 families.17 Phenotypes associated with this syndrome were first described by Mohr in 1939.18 In contrast to pathologies involving mutations in other connexins, which present with a restricted range of phenotypic abnormalities, ODDD is a pleiotropic condition that corresponds well with the wide expression pattern of Cx43 both developmentally and in mature tissue. Currently, 36 different Cx43 mutations have been identified resulting in phenotypes that range from syndactyly type III alone (mutation G143S) to a recessive mutation (R76H) displaying a phenotype characteristic of both ODDD and HallermaneCStreiff syndromes.19,20,21,22,23 Most of these occur in the N-terminal half of the Cx43 gene and affect amino acids that are highly conserved in Cx43 across various species as well as other human connexin isoforms.
The initial characterization of the cellular phenotypes associated with ODDD-relevant Cx43 mutations were presented by Seki et al.24 This study examined 3 ODDD mutations found within the cytoplasmic loop of Cx43: I130T, K134E, and G138R. The results demonstrated that all mutants were capable of forming gap junction plaques, but their ability to form functional channels was significantly compromised. Cell pairs expressing mutant G138R were not electrically coupled. Single gap junction channels were occasionally recorded from cell pairs expressing either I130T or K134E. Unitary conductance was unaffected by the I130T mutation, whereas a decrease in unitary conductance was observed in channels harboring mutation K134E.24 The study of Seki et al was followed by that of Roscoe et al,25 who confirmed our initial observations on mutation G138R and, in addition, characterized the phenotype of mutation G21R. These authors further showed that mutations G138R and G21R act as dominant negatives when coexpressed with wild-type Cx43. Here, we present the first comprehensive characterization of 8 ODDD-relevant Cx43 mutations (Figure 1) in terms of protein expression, cellular localization, and coupling properties. Mutations were selected to represent the variety of connexin protein domains affected as well as a range of clinical phenotypic consequences (see Table 1). Mutant Cx43 genes were transiently transfected in communication-deficient cells and the presence, location, and function of Cx43 was tested. Our studies show that, consistent with the wide spectrum of clinical manifestations, ODDD-relevant mutations lead to a wide range of molecular phenotypes ranging from absence of gap junctional plaques to changes in permeability of current-passing channels.
Materials and Methods
Molecular Cloning of Human GJA1 and Generation of ODDD Mutations
Molecular cloning of human Cx43 and generation of ODDD mutations. A product consisting of 201 bp of the 5' untranslated region and the coding region of the wild-type GJA1 gene sequence was amplified and cloned into vectors pEYFP-N1 or pIRES2-Ds-Red (BD Biosciences Clontech, Palo Alto, Calif) using adapter primers (forward with an Nhe I recognition site 5'-ATGCTAGCAAGCTTTTACGAGGTATCAG-3'and reverse with a Sac II recognition site without or with stop codon 5'-AGCCGCGGGATCTCCAGGTCATCAGG -3' or 5'-AACCGCGGCTAGATCTCCAGGTCATCAG- G-3'). Two-stage mutagenesis was performed using the above wild-type plasmid as a template to introduce 8 different ODDD related mutations (Figure 1) into the GJA1 sequence. Plasmid inserts were sequenced to confirm wild-type and altered GJA1.
Transient Expression of Wild-Type and Mutant Cx43 Proteins In Hela and N2A Cells
HeLa and N2A cells were maintained in DMEM (Cellgro) supplemented with 10% fetal bovine serum and 50 U/mL penicillin-streptomycin. Cell cultures were maintained at 37°C in a humidified 5% CO2 incubator. pEYFP-N1 (Clontech/BD Biosciences) vectors containing DNA coding for wild-type or mutant GJA1 DNA (0.25 to 1.0 e) were transfected into HeLa or N2A cells using an Effectene kit (Qiagen). HeLa cells were primarily used for immunolocalization studies and N2A cells for electrophysiological experiments.
Immunoblot Analysis of Cx43 Protein Expression
Western blot analysis was performed on HeLa cells 24 hours after transient transfection with wild-type or mutant Cx43. Cells were washed in 1 mL cold PBS (pH 7.4) and scraped and spun down by centrifugation at 14 000g for 5 minutes. Cells were resuspended and permeabilized with an extraction buffer containing: 150 mmol/L NaCl, 50 mmol/L Tris (pH 8.0), 0.02% sodium azide, 1 e/mL aprotonoin (Sigma), Complete protease inhibitor (Roche), and 1% Triton X-100. After a 30 second sonication, 1% PMSF was added to cell lysates followed by a 30 minute incubation on ice. Following incubation, cell lysates were centrifuged for 10 minutes at 14 000g at 4°C. The collected supernatants were then fractionated on a 4% to 8% SDS-polyacrylamide gel. Proteins from the gel were then transferred to a nitrocellulose blot, followed by a 1 hour block at room temperature in PBS containing 5% nonfat milk with 0.1% Tween. The membrane was then incubated with mouse monoclonal antibody specific for GFP (Sigma) overnight at 4°C. After washing, the membrane was incubated with goat anti-mouse HRP linked secondary antibody. Immunoreactive protein on the membrane was then visualized using an enhanced chemiluminescence protocol (ECL Kit, Amersham).
Immunochemistry of Cultured HeLa Cells
Yellow fluorescent protein (YFP)-labeled Cx43 proteins were visualized using fluorescent microscopy 24 hours after transfection into HeLa cells. Briefly, HeLa cells were grown up on coverslips and transiently transfected with Cx43WT or the relevant ODDD Cx43 mutation using an Effectene kit (Qiagen). Twenty-four hours after transfection cells were fixed in 4% paraformaldehyde for 10 to 30 minutes followed by three 5-minute washes in PBS. Coverslips were mounted onto microscope slides using Vectashield mounting medium to inhibit bleaching of the fluorescent signal. YFP-conjugated Cx43 protein was directly visualized using an Olympus IX70 microscope equipped with epifluorescence. For colocalization experiments with the ER, HeLa cells were additionally cotransfected with pDsRed2-ER Vector (BD Biosciences) encoding a fusion of red fluorescent protein, the endoplasmic reticulum (ER) targeting sequence of calreticulin and the ER retention sequence KDEL.
Electrophysiology
Electrophysiological recordings of gap junction currents were obtained from N2A cell pairs transiently transfected with a specific construct. The external recording solution contained (in mmol/L) 160 NaCl, 10 CsCl, 2 CaCl2, 0.6 MgCl2, and 10 HEPES, at pH7.4. The intracellular (pipette) solution contained (in mmol/L) 139 CsCl2, 0.5 CaCl2, 10 HEPES, 10 EGTA, 2 Na2ATP, 3 MgATP (added daily). An octanol stock solution (1.5 to 2.5 mmol) was prepared from which aliquots were taken and added directly to the external solution during the experiment. Chemical compounds for cell culture and for the preparation of the stock solutions were purchased from Sigma Chemical Company (St. Louis, Mo). Recordings were obtained using the dual patch clamp method. Both cells in the pair (cell 1 and cell 2) were independently voltage clamped at the same holding potential (eC40 mV). A series of 60 mV voltage steps were applied to cell 1 (from eC40 mV to +20 mV; 10 second duration) and junctional currents were recorded from the amplifier holding the voltage in cell 2. Most determinations of unitary conductance were obtained in the presence of 1.5 to 2.5 mmol/L octanol.
Analog signals were digitized using a 4-channel SCSI based data acquisition system (CDAT4; Cugis technology, Delaware Water Gap, Pa). Analysis of data were performed using the pClamp suite of programs (version 8.2; Axon Instruments) installed on a Pentium based computer. Current signals were filtered at 100 Hz and 1 kHz, for macroscopic and unitary conductances, respectively (eC3 db, 4-pole Bessel filter, LPF 202; Warner Instrument Corp., Hamden, Conn), and digitally sampled at 2 kHz. For single channel analysis, Axoscope or Clamplex software (pClamp version 8.2; Axon Instruments) was used. Each mirrored channel event was measured manually. The conductance was calculated by dividing the amplitude of the junctional current over the driving force (+60 mV). All-points histograms of digitized current traces as well as frequency (% of total events) distribution histograms were constructed using Origin (version 7.0; Microcal, Northampton, Mass).
Dye Transfer Analysis
Permeability of the heteromeric channels was assessed simultaneously with estimations of junctional conductance by loading 1 patch pipette with Lucifer yellow (5% in LiCl) while recording in the dual patch clamp configuration. Transfer of dye between cells expressing the relevant WT or mutant DNA was visualized using a Yellow GFP filter set (Chroma Tech.) on a Nikon Diaphot 200 microscope.
Results
Cellular Localization of ODDD Cx43 Homotypic Mutant Proteins
In order to determine whether selected ODDD mutations affected Cx43 protein expression, we performed a western blot analysis of HeLa cells transiently transfected with YFP-labeled WT or mutant Cx43 (Figure 2). All ODDD Cx43 mutant proteins were detected by antibodies specific for the YFP label as bands of the appropriate size. In order to determine whether the mutant connexin was being properly targeted to the membranes of apposed cells, Cx43 constructs concatenated with YFP were expressed in HeLa cells and visualized using epifluorescence microscopy (Figure 3). Probability of plaque formation was quantified relative to the frequency of plaques expressed by the wild-type construct (see Table 2). Significant numbers of putative gap junction plaques were observed in cells expressing Cx43WT (Figure 3A), mutations L90V, and Y17S (Figure 3B,C) and to a lesser degree (Table 2) in cells expressing A40V, I130T, G21R, and K134E (Figure 3 D-G). Cells expressing mutations F52dup and R202H (Figure 3 H,I) did not form similar plaque formations and the protein appeared to be localized in the intracellular space (see also Table 2).
To further examine the pattern of intracellular localization seen for mutant R202H, HeLa cells were cotransfected with YFP-conjugated R202H and a DsRed-conjugated, ER-specific marker (Figure 4 A through C). The results suggest that Cx43 proteins with the R202H mutation predominantly localize at the ER. Experiments with F52dup, a second mutation incapable of forming homotypic junctional plaques, also demonstrated substantial colocalization of the mutant protein with the ER marker (Figure 4 D through F).
Functional Consequences of Cells Homotypically Expressing Cx43 Mutations
In order to evaluate the ability of ODDD Cx43 mutations to form functional channels, we used the dual patch clamp method. YFP-tagged Cx43 constructs were expressed in communication deficient neuroblastoma (N2A) cells and junctional currents were recorded from homotypically-paired cells 24 to 48 hours post-transfection. The results are summarized in Table 3. Of the 8 mutations tested only 3 (L90V, I130T, and K134E) were able to induce electrical coupling between paired cells but did so at significantly lower levels compared with homotypic Cx43WT cell pairs. Furthermore, L90V and I130T displayed unitary conductances similar to those of the wild-type channel. Mutations A40V, G21R, and Y17S displayed a complete absence of electrical coupling although experimental cell pairs demonstrated plaque formations that were qualitatively similar to those seen for Cx43WT-expressing cell pairs. As previously mentioned mutations R202H and F52dup did not form gap junction plaques when expressed in HeLa cells (Figure 3). Similarly when these mutations were expressed in N2A cells they were not capable of forming electrically coupled cell pairs.
Functional Analysis of Heteromeric Cell Pairs Coexpressing Selected ODDD Mutations With Cx43WT
A majority of the ODDD patients are heterozygous, presenting only 1 mutant allele. Under these circumstances, Cx43WT may coassemble with the mutant subunits to form heteromeric channels. The diversity of these combinations can influence the nature of the cell-to-cell communication.
As mentioned previously, mutant L90V was capable of forming functional homomeric channels although macroscopic junctional conductance was significantly reduced when compared with paired cells expressing Cx43WT (Table 3). On the other hand, the rate of plaque formation for mutant L90V was approximately half that of the wild-type protein (Table 2), thus facilitating its electrophysiological characterization. Because of the combination of high-rate plaque formation but low electrical coupling, we chose L90V to examine the extent of electrical coupling in cell pairs expressing both the mutant and the wild-type protein. N2A cells were cotransfected with different ratios of mutant to wild-type Cx43 DNA while keeping the total amount of transfected DNA constant (0.5 e). When cells were transfected with an equal amount of mutant to Cx43WT cDNA, there was a significant decrease in average junctional conductance suggesting that L90V functions as a dominant negative in heterologous channels (Figure 5). When the ratio of the L90V mutant cDNA was doubled relative to that of Cx43WT the inhibition of junctional currents was even more profound. Unitary conductances of the L90V homomeric channel were similar to Cx43WT (Table 3).
Similar experiments were performed with mutations F52dup and R202H. Neither of these 2 Cx43 proteins were able to form homotypic junctional plaques and therefore, did not result in paired cells that were electrically coupled. Surprisingly, coexpression of YFP-labeled F52dup or R202H with unlabeled Cx43WT resulted in junctional plaques positive for YFP indicating that these mutant proteins localize to the cell membrane in the presence of Cx43WT and are incorporated into heteromeric gap junction channels (Figures 6 and 7).
Electrophysiological recordings demonstrated that cells cotransfected with F52dup and Cx43WT DNA, at ratios of up to 1:8 (wt:mutant) presented junctional conductance values not significantly different from those observed in homologous wild-type channels. Even when the amount of transfected cDNA for the wild-type protein was one-twentieth of that of mutant, plaque formation and electrical coupling was still detected, though junctional conductance values were significantly reduced. These data strongly suggest that wild-type subunits are able to heteromerize with the mutant connexins and by doing so, rescue their ability to participate in a functional gap junction plaque.
Similar results were found for cells coexpressing mutation R202H with Cx43WT (Figure 7). In fact, heterologous cells transfected with a wild-type:mutant DNA ratio of 1:8 still demonstrated a macroscopic junctional conductance that was not significantly different from cells homotypically expressing Cx43WT. These data are in stark contrast to the dominant negative effect observed for mutation L90V and suggest a wide spectrum of functional molecular phenotypes depending on the specific mutations in the Cx43 molecule.
Single Channel Analysis of Heteromeric Pairs
In order to determine whether the F52dup or R202H mutations affect the conductance states of heteromeric channels, we cotransfected either F52dup or R202H with Cx43WT into N2A cells at a 4:1 ratio of mutant to wild-type DNA. At these ratios there is a very high probability that the active channels contain at least 1 mutant subunit per connexon. Figure 8 shows representative traces recorded from N2A cells expressing F52dup/ Cx43WT (top) or R202H/ Cx43WT (bottom) channels accompanied by their all-points histograms. Both records demonstrate main (105 pS) and subconductance (25 pS) states that are not significantly different from what has been observed for Cx43WT homomeric channels.30
Tracer Coupling in Heteromeric Pairs
We have shown that the junctional conductance of heteromeric channels formed by coexpression of mutant F52dup or R202H with Cx43WT was not different from that observed in cell pairs expressing homologous Cx43WT. As a next step, we asked whether these mutations affected the permeability properties of the heterologous channels. For each set of experiments, N2A cells were transiently transfected with a total of 0.5 e of the relevant DNA. For cells coexpressing mutations F52dup or R202H with Cx43WT the ratio of wild-type to mutant DNA was 1:20 and 1:8, respectively. Previous experiments show that cells transfected with the pertinent DNAs at these ratios were electrically coupled (see Figures 6 and 7 ). Given the large proportion of mutant over wild-type DNA, the largest fraction of subunits within each hexamer is likely to come from the mutant protein. For these experiments, one of the patch pipettes was loaded with Lucifer yellow, a negatively charged fluorescent dye; junctional conductance was measured by dual patch clamp and cell-to-cell transfer of dye was simultaneously monitored. Our results show that 30 minutes after Lucifer yellow was introduced into the donor cell of a homotypic Cx43 pair (Figure 9 top) there was appreciable dye transfer to the recipient cell. Similarly, heteromeric channels formed by coexpressing Cx43WT with mutation F52dup showed significant dye transfer after 30 minutes (Figure 9 middle). In contrast heteromeric channels coexpressing Cx43WT and R202H did not transfer dye (Figure 9 bottom). Junctional conductance for the Cx43WT, Cx43WT/F52dup, and Cx43WT/R202H experiments shown were 16.7, 2.4, and 6.7 nS, respectively. These findings, together with those presented in Figures 7 and 8, show that the R202H mutation altered the permeability of Cx43 gap junctions though electrical coupling was unaffected by the mutation.
Discussion
In this study we have characterized the consequences of 8 different ODDD Cx43 mutations on the expression and function of the Cx43 protein. All mutations were able to generate protein products of the appropriate size although their ability to assemble into functional gap junction plaques varied considerably. Mutations F52dup and R202H, located in the first and second extracellular loops (EL1 and EL2), respectively, were not able to form homomeric junctional plaques; the Cx43-immunoreactive protein was localized to the intracellular space and primarily associated with the ER. Mutations in the N-terminus (Y17S) and first transmembrane domain (G21R, A40V) formed gap junction plaques although homotypically paired cells were not electrically coupled. Finally, mutations L90V, I130T, and K134E did form electrically coupled gap junctions with mutants L90V and I130T demonstrating unitary conductances similar to Cx43WT. However, macroscopic junctional conductances were significantly reduced for all 3 of these mutations relative to wild-type cell pairs.
Three of the mutations studied (Y17S, G21R, and A40V) formed the appropriate protein product and were transported to the membrane to form apparent plaque-like structures but did not form functional channels when paired homotypically (Table 3). It is our interpretation that these mutations were able to form gap junction plaques and yet the channels remained in a closed conformation. An alternative explanation is that the apparent YFP positive plaques are composed of gap junctions internalized into vesicular-like structures referred to as "annular gap junctions" and remain in close proximity to regions of cell-to-cell contact.31 Furthermore, the possibility that the mutant protein aggregated beneath the membrane without forming a mature gap junction plaque, though improbable, cannot be discarded. It is interesting to note that these 3 mutations are found in the N-terminal (NT) and first transmembrane (TM1) domains, and both of these regions have been shown to play a critical role in influencing the voltage-dependent gating of the Cx43 channel. Studies in Cx32 have shown that amino acids located in the NT and the border of TM1/EL1 work cooperatively to form a charge complex that functions as a critical part of the channel voltage sensor.32,33 Residue A40 is located near the TM1-EL1 interface. Though the A40V mutation is conserved in terms of charge, there is a significant difference in the Van der Waals volume occupied by these residues,34 which could destabilize the charge complex and modify channel open probability. A net charge effect is more likely for the substitution G21R, which could potentially modify the spatial arrangement of the first transmembrane domain and force the channel into a closed configuration. Together with Y17S, these mutations may prevent functional channel formation by altering the structural determinants that control channel gating, thus forcing the pore into a permanent closed position.
Mutations F52dup and R202H failed to form gap junction plaques. These mutations are located in the extracellular domains (EL1 and EL2, respectively). The primary structure of domains EL1 and EL2 is highly conserved among the members of the connexin family. In particular, there is conservation of 6 cysteine residues thought to be essential for connexin stabilization, connexin formation, and connexon docking.35,36 Interestingly, the F52dup and R202H mutations are located in proximity to cysteine residues C54 and C198, respectively. It is tempting to speculate that these mutations prevent proper formation of connexons and their subsequent trafficking to the plasma membrane. In agreement with our data, it has been shown that point mutations in conserved residues found in EL2 of Cx43 (F199L, R202E, and E205H) prevent localization of the Cx43 protein to the plasma membrane.37 Also consistent with our results, the investigators were able to rescue 2 of the mutant proteins (F199L and R202E) to sites of cell-to-cell contact by coexpression with Cx43WT.
Mutants L90V, I130T, and K134E formed electrically capable homotypic gap junction channels, though their probability of plaque formation was lower to that of the wild-type protein and macroscopic conductance was consistently lower than that of wild-type-expressing cells. Other electrical parameters such as voltage dependence or the characteristics of the residual state could not be evaluated because they were likely affected by the YFP concatenation (see, eg, the work of Bukauskas et al on Cx43GFP concatenants38). Unfortunately, given the low probability of plaque formation, the fluorescent tag was required to identify the few cell pairs expressing gap junction channels, thus making it impractical to characterize the properties of untagged Cx43 channels.
A majority of the ODDD patients are heterozygotes, presenting only 1 mutant allele. It is therefore possible that, unless a specific mutation interferes with connexin oligomerization, the majority of Cx43 gap junctions in ODDD patients are heterologous combinations of wild-type and mutant proteins. The behavior of these heterologous channels may vary depending on the specific mutation. A correlation between the cellular channelopathy and the clinical expression of the mutations requires a thorough characterization of the function of the mutant protein when coexpressed with wild-type. Here, we have looked at the behavior of 3 of the mutations. Functional conductance between cells expressing Cx43WT was significantly decreased after coexpression of mutant L90V. On the other hand, the lack of plaque formation observed in cells expressing homologous R202H and F52dup proteins was readily rescued by coexpression of the wild-type protein. Moreover, the function of the heterologous R202H channel was not the same as that of wild-type. Indeed, the heterologous channels were no longer permeable to the negatively charged fluorescent dye Lucifer yellow (Figure 9). This provides another line of evidence that the R202H subunit has been integrated into heterologous functional channels. Furthermore, the data suggest that although the constrictive barrier for electrical conduction was not affected, permselectivity was compromised by the mutation.
The results obtained from coexpression of mutants R202H and F52dup with Cx43WT can be explained by the formation of heteromeric connexons and the consequent incorporation of those heteromers into gap junction plaques with specific properties. However, it is indeed possible that all channels are homomeric and junctional plaques formed by a combined population of 2 types of homologous channels. Yet, we find this possibility somewhat improbable. Indeed, the latter would imply that homomeric connexons of one type are able to incite the trafficking of (and incorporate into apparent plaques) other homomeric connexons that normally do not traffic to the plasma membrane. This assumes an intermolecular interaction between connexons either direct, or mediated by a separate molecule. We are not aware of studies indicating "passive trafficking" of one oligomer as it is carried by the other. Moreover, it is worth noting that the channels resulting from the coexpression of R202H with Cx43WT are electrically conductive yet impermeable to Lucifer Yellow. The suggestion that both channels are homomeric would then assume that the homomeric R202H connexons are capable of modifying the permeability of homomeric Cx43WT connexons. This implies a long-range interaction between connexons that has not been described and seems very improbable given the intermolecular distances. Thus, the most likely scenario is that R202H (and, likely, F52dup) are rescued via oligomerization with Cx43WT subunits. Yet the possibility that the observed function results from the coexistence of homomeric connexons, though unlikely, cannot be completely discarded.
Similarly, a recent study examining 2 ODDD mutations, G21R and G138R, revealed that both mutants significantly inhibited dye coupling by Lucifer yellow when expressed in neuronal rat kidney (NRK) cells by hampering endogenous Cx43 function.25 These types of functional changes may interfere with the ability of molecular messages to propagate between cells and coordinate cell function. Similar experiments performed for F52dup/Cx43WT channels showed that unlike the R202H heteromerics, permeability to Lucifer yellow in these channels remained unaltered. From that perspective the specific liability of this mutation to Cx43 gap junctions remains to be determined.
Previous studies have shown that Cx43-null mice develop a severe cardiac malformation.1 In addition, conditional Cx43KO mice are very susceptible to lethal cardiac arrhythmias.39 In that context, the lack of a consistent cardiac phenotype in ODDD patients may seem surprising at first glance. However, it is important to note that the presence of a single Cx43 allele is enough to maintain normal cardiogenesis and normal electrical synchrony9 in Cx43-deficient mice. Moreover, it should be noted that in most ODDD mutations, the integrity of the CT domain is preserved. As such, these proteins are able to maintain normal protein-protein interactions, thus preserving the molecular integrity of the gap junction plaque. The relevance of this aspect of connexin function is emphasized by the recent study of Maass et al showing that mice expressing a mutant Cx43 protein lacking most of the CT domain present a lethal skin phenotype because of defects in the epidermal barrier.40 Interestingly, Palmoplantar Keratoderma and other epidermal defects have been identified in patients with a Cx43 frameshift mutation consequent to a 2 base pair deletion in the C-terminal domain. The frameshift resulted in a stretch of 46 incorrect amino acids and the premature termination of the cytoplasmic terminal.41 Coincidentally both the frameshift and the K134E mutation (which is also associated with skin defects) involve cytoplasmic regions of Cx43 that have been shown to be involved in direct intramolecular interactions essential for molecular dynamics of connexin regulation.42
Correlating the functional effects of each mutation to the manifestations of the disease is complicated by the inherent variability of associated symptoms both within families carrying the same mutation as well as across families expressing different mutations. The wide variety of clinical manifestations (ocular, dental, digital, and craniofacial anomalies with secondary symptoms involving neurological and cardiac abnormalities, conductive hearing loss, cataracts, and glaucoma (14; see also Table 1) coincides with the wide expression of Cx43 both in development and after birth.43 Interestingly, whereas ODDD is associated with a broad spectrum of neurological symptoms, cardiac abnormalities remain somewhat rare.44 The latter contrasts with the critical role Cx43 plays in both cardiogenesis and electrical synchrony.45,46 An increase in the clinical registry will provide more data to further assess this hypothesis.
In this study we have presented data on 8 different ODDD mutations that affect Cx43 gap junctions in manners ranging from the failure of the mutant protein to reach cell membrane to the formation of functional gap junction channels with altered biophysical characteristics. A preponderance of the data suggests that the mechanism by which these mutations contribute to the myriad of symptoms associated with the ODDD syndrome is by reducing Cx43 cell-to-cell communication. This in turn could lead to disruptions in morphological patterning during development and the disruption of normal cell-to-cell coupling in mature tissue. Future studies of these naturally occurring Cx43 mutations will contribute to a better understanding of the mechanisms responsible for ODDD and the structural bases of Cx43 function.
Acknowledgments
This work was supported by grants from the National Institutes of Health P01-HL-39707, R01-GM-5769, and NIH DE 13849.
References
Reaume AG, de Sousa PA, Kulkarni S, Langille BL, Zhu D, Davies TC, Juneja SC, Kidder GM, Rossant J. Cardiac malformation in neonatal mice lacking connexin43. Science. 1995; 267: 1831eC1834.
Lo CW, Wessels A. Cx43 gap junctions in cardiac development. Trends Cardiovasc Med. 1998; 8: 264eC269.
Lowenstein WR, Kanno Y. The cell-cell in growth control. Sem. Cell. Bio. 149: 295eC298.
Saez JC, Connor JA, Spray DC, Bennett MV. Hepatocyte gap junctions are permeable to the second messenger, inositol 1,4,5-trisphosphate, and to calcium ions. Proc Natl Acad Sci U S A. 1989; 86: 2708eC2712.
Moorby C, Patel M. Dual functions for connexins: Cx43 regulates growth independently of gap junction formation. Exp Cell Res. 2001; 271: 238eC248.
Willecke K, Eiberger J, Degen J, Eckardt D, Romualdi A, Guldenagel M, Deutsch U, Sohl G. Structural and functional diversity of connexin genes in the mouse and human genome. Biol Chem. 2002; 383: 725eC737.Review.
Beyer EC, Paul DL, Goodenough DA. Connexin family of gap junction proteins. J Membr Biol. 1990; 116: 187eC194.
Beauchamp P, Choby C, Desplantez T, de Peyer K, Green K, Yamada KA, Weingart R, Saffitz JE, Kleber AG. Electrical propagation in synthetic ventricular myocyte strands from germline connexin43 knockout mice. Circ Res. 2004; 95: 170eC178.
Morley GE, Vaidya D, Samie FH, Lo C, Delmar M, Jalife J. Characterization of conduction in the ventricles of normal and heterozygous Cx43 knockout mice using optical mapping. J. Cardiovasc Electrophysiol. 1999: 1361eC1375.
Smith JH, Green CR, Peters NS, Rothery S, Severs NJ. Altered patterns of gap junction distribution in ischemic heart disease. An immunohistochemical study of human myocardium using laser scanning confocal microscopy. Am J Pathol. 1991; 139: 801eC821.
Dupont E, Matsushita T, Kaba RA, Vozzi C, Coppen SR, Khan N, Kaprielian R, Yacoub MH, Severs NJ. Altered connexin expression in human congestive heart failure. J Mol Cell Cardiol. 2001; 33: 359eC371.
Emdad L, Uzzaman M, Takagishi Y, Honjo H, Uchida T, Severs NJ, Kodama I, Murata Y. Gap junction remodeling in hypertrophied left ventricles of aortic-banded rats: prevention by angiotensin II type 1 receptor blockade. J Mol Cell Cardiol. 2001; 33: 219eC231.
Kelsell DP, Dunlop J, Stevens HP, Lench NJ, Liang JN, Parry G, Mueller RF, Leigh IM. Connexin 26 mutations in hereditary non-syndromic sensorineural deafness. Nature. 1997; 387: 80eC83.
Richard G. Connexin disorders of the skin. Adv Dermetol. 2001; 17: 243eC277.
Nakase T, Naus CG. Gap junctions and neurological disorders of the central nervous system. Biochim Biophys Acta. 2004; 1662: 149eC158.
Bergoffen J, Scherer SS, Wang S, Scott MO, Bone LJ, Paul DL, Chen K, Lensch MW, Chance PF, Fischbeck KH. Connexin mutations in X-linked Charcot-Marie-Tooth disease. Science. 1993; 262: 2039eC2042.
Paznekas WA, Boyadjiev SA, Shapiro RE, Daniels O, Wollnik B, Keegan CE, Innis JW, Dinulos MB, Christian C, Hannibal MC, Jabs EW. Connexin 43 (GJA1) mutations cause the pleiotropic phenotype of oculodentodigital dysplasia. Am J Hum Genet. 2003; 72: 408eC418.
Mohr OL. Dominant acrocephalosyndacttyly. Hereditas. 1939; 25: 193eC203.
Richardson R, Donnai D, Meire F, Dixon MJ. Expression of Gja1 correlates with the phenotype observed in oculodentodigital syndrome/type III syndactyly. J Med Genet. 2004; 41: 60eC67.
Pizzuti A, Flex E, Mingarelli R, Salpietro C, Zelante L, Dallapiccola B. A homozygous GJA1 gene mutation causes a Hallermann-Streiff/ODDD spectrum phenotype. Hum Mutat. 2004; 23: 286.
van Steensel MA, Spruijt L, van der Burgt I, Bladergroen RS, Vermeer M, Steijlen PM, van eel M. A 2-bp deletion in the GJA1 gene is associated with oculo-dento-digital dysplasia with palmoplantar keratoderma. Am J Med Genet A. 2005; 132: 171eC174.
Kjaer KW, Hansen L, Eiberg H, Leicht P, Opitz JM, Tommerup N. Novel Connexin 43 (GJA1) mutation causes oculo-dento-digital dysplasia with curly hair. Am J Med Genet. 2004; 127A: 152eC157.
Vitiello C, D’Adamo P, Gentile F, Vingolo EM, Gasparini P, Banfi, S. A novel GJA1 mutation causes oculodentodigital dysplasia without syndactyly. Am J Med Genet A. 2005; 133A: 58eC60.
Seki A, Coombs W, Taffet SM, Delmar M. Loss of electrical communication, but not plaque formation, after mutations in the cytoplasmic loop of connexin43. Heart Rhythm J. 2004; 1: 1eC5.
Roscoe W, Veitch GI, Gong XQ, Pellegrino E, Bai D, McLachlan E, Shao Q, Kidder G, Laird DW. Oculodentodigital dysplasia-causing connexin43 mutants are non-functional and exhibit dominant effects on wild-type connexin43. J Biol Chem. In press.
Rajic DS, deVeber LL. Hereditary oculodentoosseous dysplasia. Ann Radiol. 1966; 9: 224.
Gellis SS, Feingold M. Oculodentodigital dysplasia. Picture of the month. Am J Dis Child. 1974; 128: 81eC82.
Weintraub DM, Baum JL, Pashayan HM. A family with oculodentodigital dysplasia. Cleft Palate J. 1975; 12: 323eC329.
Opjordsmoen S, Nyberg-Hansen R. Hereditary spastic paraplegia with neurogenic bladder disturbances and syndactylia. Acta Neurol Scandinav. 1980; 61: 35eC41.
Valiunas V, Bukauskas FF, Weingart R. Conductances and selective permeability of connexin43 gap junction channels examined in neonatal rat heart cells. Circ Res. 1997; 80: 708eC719.
Jordan K, Chodock R, Hand AR, Laird DW. The origin of annular junctions: a mechanism of gap junction internalization. J Cell Sci. 2001; 114: 763eC773.
Verselis VK, Ginter CS, Bargiello TA. Opposite voltage gating polarities of two closely related connexins. Nature. 1994; 368: 348eC351.
Purnick PE, Oh S, Abrams CK, Verselis VK, Bargiello TA. Reversal of the gating polarity of gap junctions by negative charge substitutions in the N-terminus of connexin 32. Biophys J. 2000; 79: 2403eC2415.
Creighton TE. Proteins. W.H. Freeman and Company. New York, United States. 1993: 507.
Rahman S, Evans WH. Topography of connexin32 in rat liver gap junctions. Evidence for an intramolecular disulphide linkage connecting the two extracellular peptide loops. J Cell Sci. 1991; 100: 567eC578.
Foote CI, Zhou L, Zhu X, Nicholson BJ. The pattern of disulfide linkages in the extracellular loop regions of connexin 32 suggests a model for the docking interface of gap junctions. J Cell Biol. 1998; 140: 1187eC1197.
Olbina G, Eckhart W. Mutations in the second extracellular region of connexin 43 prevent localization to the plasma membrane, but do not affect its ability to suppress cell growth. Mol Cancer Res. 2003; 1: 690eC700.
Bukauskas FF, Jordan K, Bukauskiene A, Bennett MV, Lampe PD, Laird DW, Verselis VK. Clustering of connexin 43-enhanced green fluorescent protein gap junction channels and functional coupling in living cells. Proc Natl Acad Sci U S A. 2000; 97: 2556eC2561.
Vink MJ, Suadicani SO, Vieira DM, Urban-Maldonado M, Gao Y, Fishman GI, Spray DC. Alterations of intercellular communication in neonatal cardiac myocytes from connexin43 null mice. Cardiovasc Res. 2004; 62: 397eC406.
Maass K, Ghanem A, Kim JS, Saathoff M, Urschel S, Kirfel G, Grummer R, Kretz M, Lewalter T, Tiemann K, Winterhager E, Herzog V, Willecke K. Defective epidermal barrier in neonatal mice lacking the C-terminal region of connexin43. Mol Biol Cell. 2004; 15: 4597eC608.Epub 2004 Jul 28.
van Steensel MA, Spruijt L, van der Burgt I, Bladergroen RS, Vermeer M, Steijlen PM, van Geel M. A 2-bp deletion in the GJA1 gene is associated with oculo-dento-digital dysplasia with palmoplantar keratoderma. Am J Med Genet A. 2005; 132: 171eC174.
Duffy HS, Sorgen PL, Girvin ME, O’Donnell P, Coombs W, Taffet SM, Delmar M, Spray DC. pH-dependent intramolecular binding and structure involving Cx43 cytoplasmic domains. J Biol Chem. 2002; 277: 36706eC36714.
Ruangvoravat CP, Lo CW. Connexin 43 expression in the mouse embryo: localization of transcripts within developmentally significant domains. Dev Dyn. 1992; 194: 261eC281.
Loddenkemper T, Grote K, Evers S, Oelerich M, Stogbauer F. Neurological manifestations of the oculodentodigital dysplasia syndrome. J Neurol. 2002; 249: 584eC595.
Lo CW, Wessels A. Cx43 gap junctions in cardiac development. Trends Cardiovasc Med. 1998; 8: 264eC269.
Rohr, S. Role of gap junctions in the propagation of the cardiac action potential. Cardiovasc Res. 2004; 62: 309eC322.Review.