Gating of connexin 43 gap junctions by a cytoplasmic loop calmodulin binding domain

Calmodulin (CaM) binding sites were recently identified on the cytoplasmic loop (CL) of at least three α-subfamily connexins (Cx43, Cx44, Cx50), while Cx40 does not have this putative CaM binding domain. The purpose of this study was to examine the functional relevance of the putative Cx43 CaM binding site on the Ca(2+)-dependent regulation of gap junction proteins formed by Cx43 and Cx40. Dual whole cell patch-clamp experiments were performed on stable murine Neuro-2a cells expressing Cx43 or Cx40. Addition of ionomycin to increase external Ca(2+) influx reduced Cx43 gap junction conductance (G(j)) by 95%, while increasing cytosolic Ca(2+) concentration threefold. By contrast, Cx40 G(j) declined by <20%. The Ca(2+)-induced decline in Cx43 G(j) was prevented by pretreatment with calmidazolium or reversed by the addition of 10 mM EGTA to Ca(2+)-free extracellular solution, if Ca(2+) chelation was commenced before complete uncoupling, after which g(j) was only 60% recoverable. The Cx43 CL(136-158) mimetic peptide, but not the scrambled control peptide, or Ca(2+)/CaM-dependent kinase II 290-309 inhibitory peptide also prevented the Ca(2+)/CaM-dependent decline of Cx43 G(j). Cx43 gap junction channel open probability decreased to zero without reductions in the current amplitudes during external Ca(2+)/ionomycin perfusion. We conclude that Cx43 gap junctions are gated closed by a Ca(2+)/CaM-dependent mechanism involving the carboxyl-terminal quarter of the connexin CL domain. This study provides the first evidence of intrinsic differences in the Ca(2+) regulatory properties of Cx43 and Cx40.     

Identification of the calmodulin binding domain of alpha-fodrin and implications for folding

A cDNA clone producing a protein that binds calmodulin has been isolated from a mouse macrophage library. The cDNA was sequenced and identified as coding for fodrin. By deleting part of the sequence, the calmodulin binding domain was located. The site is situated on repeat 11 of fodrin probably on its extra arm. This part of the sequence exhibits great similarity to other calmodulin binding proteins. Analysis of the sequence and spatial structure of calmodulin revealed a domain which is quite complementary to the sequence identified on fodrin. These results provide a new insight into the structure of fodrin and consequently into the structure of proteins of the spectrin family. A model for the general folding of these molecules is proposed, involving a simple three-layer folding. The structure was further corroborated by analysis of charge distribution in the vicinity of the calmodulin binding site. The folding we propose is in good agreement with digestion experiments and explains observations in diseases resulting from mutations of human spectrin.     

The regulatory role of the C-terminal domain of connexin43

The C-terminal (CT) domain of connexin43 (Cx43) is thought to be important in the control of gap junction function via: a.) CT phosphorylation-dependent control of gap junction assembly and gating, b.) interactions of CT with key regulatory binding partners. To more closely examine CT-dependent regulation, we have expressed a hemagglutinin-Cx43CT (amino acids 235-382) fusion protein in Normal Rat Kidney (NRK) cells under a tetracycline-responsive inducible promoter. Western blot analysis shows that Cx43CT expression is markedly induced by at least 48 h oftreatment with the tetracycline analogue, doxycycline. Furthermore, Cx43CT is modified within the cell, as several treatments/conditions that increase endogenous Cx43 phosphorylation induced a mobility shift in Cx43CT. Treatment with kinase activators, including epidermal growth factor (EGF) and the tumor promoting phorbol ester 12-O-tetradecanylphorbol-13-acetate (TPA), caused a shift in the mobility of the Cx43CT in a manner consistent with the mobility shift observed upon increased phosphorylation of endogenous Cx43. Similarly, Cx43CT in mitotic cells is extensively shifted, consistent with reports which show that Cx43 is phosphorylated to a unique phosphoisoform in mitotic cells. These results indicate that the Cx43CT can interact with at least some of the kinases that phosphorylate endogenous Cx43 in cells and possibly modulate the effects of kinase activation on gap junctional communication.     

Identification of an atypical calcium-dependent calmodulin binding site on the C-terminal domain of GluN2A

N-methyl-d-aspartate (NMDA) receptors are calcium-permeable ion channels assembled from four subunits that each have a common membrane topology. The intracellular carboxyl terminal domain (CTD) of each subunit varies in length, is least conserved between subunits, and binds multiple intracellular proteins. We defined a region of interest in the GluN2A CTD, downstream of well-characterized membrane-proximal motifs, that shares only 29% sequence similarity with the equivalent region of GluN2B. GluN2A (amino acids 875–1029) was fused to GST and used as a bait to identify proteins from mouse brain with the potential to bind GluN2A as a function of calcium. Using mass spectrometry we identified calmodulin as a calcium-dependent GluN2A binding partner. Equilibrium fluorescence spectroscopy experiments indicate that Ca²⁺/calmodulin binds GluN2A with high affinity (5.2 ± 2.4 nM) in vitro. Direct interaction of Ca²⁺/calmodulin with GluN2A was not affected by disruption of classic sequence motifs associated with Ca²⁺/calmodulin target recognition, but was critically dependent upon Trp-1014. These findings provide new insight into the potential of Ca²⁺/calmodulin, previously considered a GluN1-binding partner, to influence NMDA receptors by direct association.     

Calcium-dependent and -independent binding of soybean calmodulin isoforms to the calmodulin binding domain of tobacco MAPK phosphatase-1

The recent finding of an interaction between calmodulin (CaM) and the tobacco mitogen-activated protein kinase phosphatase-1 (NtMKP1) establishes an important connection between Ca(2+) signaling and the MAPK cascade, two of the most important signaling pathways in plant cells. Here we have used different biophysical techniques, including fluorescence and NMR spectroscopy as well as microcalorimetry, to characterize the binding of soybean CaM isoforms, SCaM-1 and -4, to synthetic peptides derived from the CaM binding domain of NtMKP1. We find that the actual CaM binding region is shorter than what had previously been suggested. Moreover, the peptide binds to the SCaM C-terminal domain even in the absence of free Ca(2+) with the single Trp residue of the NtMKP1 peptides buried in a solvent-inaccessible hydrophobic region. In the presence of Ca(2+), the peptides bind first to the C-terminal lobe of the SCaMs with a nanomolar affinity, and at higher peptide concentrations, a second peptide binds to the N-terminal domain with lower affinity. Thermodynamic analysis demonstrates that the formation of the peptide-bound complex with the Ca(2+)-loaded SCaMs is driven by favorable binding enthalpy due to a combination of hydrophobic and electrostatic interactions. Experiments with CaM proteolytic fragments showed that the two domains bind the peptide in an independent manner. To our knowledge, this is the first report providing direct evidence for sequential binding of two identical peptides of a target protein to CaM. Discussion of the potential biological role of this interaction motif is also provided.     

Complex formation between calmodulin and a peptide from the intracellular loop of the gap junction protein connexin43: Molecular conformation and energetics of binding

Gap junctions are formed by a family of transmembrane proteins, connexins. Connexin43 is a widely studied member of the family, being ubiquitously expressed in a variety of tissues and a target of a large number of disease mutations. The intracellular loop of connexin43 has been shown to include a calmodulin binding domain, but detailed 3-dimensional data on the structure of the complex are not available. In this study, we used a synthetic peptide from this domain to reveal the conformation of the calmodulin-peptide complex by small angle X-ray scattering. Upon peptide binding, calmodulin lost its dumbbell shape, adopting a more globular conformation. We also studied the energetics of the interaction using calorimetry and computational methods. All our data indicate that calmodulin binds to the peptide from cx43 in the classical ‘collapsed’ conformation.     

Identification and primary structure of a calmodulin binding domain of the Ca2+ pump of human erythrocytes

Exposure of the purified Ca2+ pump of human erythrocytes to chymotrypsin led to the rapid loss of calmodulin activation. A fragment of about 12 kDa was removed from the ATPase in 1-2 min. Blotting experiments with 125I-labeled calmodulin showed that this fragment contains the calmodulin binding region. The remainder of the ATPase molecule was degraded to a number of fragments ranging from 3 to 120 kDa; none of them bound calmodulin. To isolate the calmodulin binding domain, calmodulin which had been coupled to the Denny-Jaffe reagent (a cleavable radioactive photoaffinity cross-linker) was allowed to bind to the Ca2+ pump. After illumination to couple the cross-linker to the pump, the cleavable bond was split and the calmodulin removed, leaving the pump radioactively labeled. This pump was digested with chymotrypsin, and the products were separated by gel permeation chromatography. The only radioactive peak (migrating at about 12 kDa) was further purified on reverse-phase high pressure liquid chromatography (HPLC). Amino acid analysis showed the fragment to have a minimal molecular mass of 12.4 kDa and to contain a single methionine. After attempts to sequence the peptide directly failed. CNBr digestion was carried out on the labeled ATPase, producing both soluble and insoluble labeled material. After reverse-phase HPLC purification of the soluble material, a single radioactive peak was collected. Its sequence was (Formula: see text). A portion of this peak was passed through a microcalmodulin column; it bound in the presence of Ca2+ and was eluted by EDTA, and by a mixture of EDTA and urea. Staphylococcal V8 protease digestion of the eluted peak produced the same sequence as shown above, but starting at Leu-2 and ending at Glu-32. Structural analysis of this peptide showed that it shares features with the calmodulin binding domains of other enzymes which are regulated by calmodulin.     

Tryptophan fluorescence of calmodulin binding domain peptides interacting with calmodulin containing unnatural methionine analogues

The interactions between the abundant methionine residues of the calcium regulatory protein calmodulin (CaM) and several of its binding targets were probed using fluorescence spectroscopy. Tryptophan steady-state fluorescence from peptides encompassing the CaM-binding domains of the target proteins myosin light chain kinase (MLCK), cyclic nucleotide phosphodiesterase (PDE) and caldesmon site A and B (CaD A, CaD B), and the model peptide melittin showed Ca(2+)-dependent blue-shifts in their maximum emission wavelength when complexed with wild-type CaM. Blue-shifts were also observed for complexes in which the CaM methionine residues were replaced by selenomethionine, norleucine and ethionine, and when a quadruple methionine to leucine C-terminal mutant of CaM was studied. Quenching of the tryptophan fluorescence intensity was observed with selenomethionine, but not with norleucine or ethionine substituted protein. Fluorescence quenching studies with added potassium iodide (KI) demonstrate that the non-native proteins limit the solvent accessibility of the Trp in the MLCK peptide to levels close to that of the wild-type CaM-MLCK interaction. Our results show that the methionine residues from CaM are highly sensitive to the target peptide in question, confirming the importance of their role in binding interactions. In addition, we provide evidence that the nature of binding in the CaM-CaD B complex is unique compared with the other complexes studied, as the Trp residue of this peptide remains partially solvent exposed upon binding to CaM.     

Identification of rab20 as a potential regulator of connexin 43 trafficking

Connexin oligomerization and trafficking are regulated processes. To identify proteins that control connexin 43 (Cx43), a screen was designed using HeLa cells expressing a Cx43 construct with di-lysine endoplasmic reticulum (ER)-retention/retrieval motif, Cx43-HKKSL. At moderate levels of expression, Cx43-HKKSL is retained in the ER as monomers; however, Cx43-HKKSL stably overexpressed by HeLa cells localizes to the perinuclear region and oligomerizes. HeLa/Cx43-HKKSL overexpressors were transiently transfected with pooled clones from a human kidney cDNA library and used immunofluorescence microscopy to identify cDNAs that enabled overexpressed Cx43-HKKSL to convert from a perinuclear to ER localization pattern. Using this approach, a small molecular weight GTPase, rab20, was identified as a candidate protein with the ability to regulate Cx43 trafficking. Enhanced green fluorescent protein (EGFP)-tagged rab20 showed a predominantly perinuclear and ER localization pattern and caused wild-type Cx43 to be retained inside the cell. By contrast, mutant EGFP-rab20T19N, which lacks the ability to bind GTP, had no effect on Cx43. These results suggest Cx43 is transported through an intracellular compartment regulated by rab20 along the secretory pathway.     

Immunolocalization of an extracellular domain of connexin43 in rat heart gap junctions.

Antipeptide antibodies directed to residues 55 to 66 (NTQQPGCENVCY) of connexin43 (cx43) specifically recognize this protein on Western blots of intact and urea-split gap junctions isolated from rat heart. These antibodies detect a single protein of 43 kDa, corresponding to cx43, on Western blots of whole fractions of various vertebrate hearts. Immunogold labeling by electron microscopy shows that the epitopes recognized by these antibodies are not localized on the cytoplasmic surfaces of intact gap junctions but only at the edges of these junctions. In urea-split gap junctions the gold particles are seen in the junctional space, associated with the extracellular faces of junctional membranes. Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) analyses of rat heart gap junctions treated with trypsin show that they are constituted with either two polypeptides of Mr 12,000 and 14,000 or a single polypeptide of Mr 22,000 according to whether the analyses are performed under reducing or non-reducing conditions, respectively. The antibodies directed to residues 55 to 66 of cx43 cross-react with both the 12 and 22 kDa polypeptides. These results suggest that the two protected domains of 12 and 14 kDa which contain the first extracellular loop and a putative second extracellular loop, respectively, are linked by disulfide bonds. In adult rat heart sections analyzed by indirect immunofluorescence the intercalated discs are labeled with antibodies directed to a cytoplasmic carboxy-terminal domain of cx43 (El Aoumari et al., J. Membr. Biol. 115, 229-240 (1990)). The same intercalated discs are also labeled in adjacent sections incubated with the antibodies directed to residues 55 to 66. Two hypotheses might explain these results: either the antibodies have access to the extracellular domain of cx43 molecules localized at the edges of the gap junctions, or cx43 molecules are present in the non-junctional membranes of the intercalated discs.     

Trifluoperazine binding to calmodulin: a shift reagent 43CA NMR study

43Ca NMR experiments of Ca2+ binding to calmodulin (CaM) were performed in the presence and absence of the calmodulin antagonist trifluoperazine (TFP). By making use of the shift reagent Dy(PPP)(7-) (a 1:2 complex of DyCl3 and Na5P3O10) we have succeeded in separating the 43Ca resonances of protein-bound Ca2+ and free Ca2+ in the otherwise unresolved spectra. This experimental strategy has allowed us to demonstrate unequivocally that the affinity of CaM for Ca2+ is markedly increased in the presence of TFP. Thus Ca2+ is not liberated from the protein upon addition of TFP as had been suggested based on earlier 43Ca NMR experiments (Shimuzu, T., Hatano, M., Nagao, S. and Nozawa, Y. (1982), Biochem. Biophys. Res. Comm. 106, 1112-1118).     

A calmodulin and alpha-subunit binding domain in human erythrocyte spectrin

Human erythrocyte spectrin binds calmodulin weakly under native conditions. This binding is enhanced in the presence of urea. The site responsible for this enhanced binding in urea has now been shown to reside in a specific region of the spectrin beta-subunit. Cleavage of spectrin with trypsin, cyanogen bromide or 2-nitro-5-thiocyanobenzoic acid generates fragments of the molecule which retain the ability to bind calmodulin under denaturing conditions. The origin of these fragments, identified by two-dimensional peptide mapping, is the terminal region of the spectrin beta-IV domain. The smallest peptide active in calmodulin binding is a 10 000 Mr fragment generated by cyanogen bromide cleavage. Only the intact 74 000 Mr fragment generated by trypsin (the complete beta-IV domain) retains the capacity to reassociate with the isolated alpha-subunit of spectrin. The position of a putative calmodulin binding site near a site for subunit-subunit association and protein 4.1 and actin binding suggests a possible role in vivo for calmodulin regulation of the spectrin-actin membrane skeleton or for regulation of subunit-subunit associations. This beta-subunit binding site in erythrocyte spectrin is found in a region near the NH2-terminus at a position analogous to the alpha-subunit calmodulin binding site previously identified in a non-erythroid spectrin by ultrastructural studies.     

Binding of calcium by calmodulin: influence of the calmodulin binding domain of the plasma membrane calcium pump.

The interaction between calmodulin and synthetic peptides corresponding to the calmodulin binding domain of the plasma membrane Ca2+ pump has been studied by measuring Ca2+ binding to calmodulin. The largest peptide (C28W) corresponding to the complete 28 amino acid calmodulin binding domain enhanced the Ca2+ affinity of calmodulin by more than 100 times, implying that the binding of Ca2+ increased the affinity of calmodulin for the peptide by more than 10(8) times. Deletion of the 8 C-terminal residues from peptide C28W did not decrease the affinity of Ca2+ for the high-affinity sites of calmodulin, but it decreased that for the low-affinity sites. A larger deletion (13 residues) decreased the affinity of Ca2+ for the high-affinity sites as well. The data suggest that the middle portion of peptide C28W interacts with the C-terminal half of calmodulin. Addition of the peptides to a mixture of tryptic fragments corresponding to the N- and C-terminal halves of calmodulin produced a biphasic Ca2+ binding curve, and the effect of peptides was different from that on calmodulin. The result shows that one molecule of peptide C28W binds both calmodulin fragments. Interaction of the two domains of calmodulin through the central helix is necessary for the high-affinity binding of four Ca2+ molecules.     

Intracellular calcium regulation of connexin43

The mechanism by which intracellular Ca(2+) concentration ([Ca(2+)](i)) regulates the permeability of gap junctions composed of connexin43 (Cx43) was investigated in HeLa cells stably transfected with this connexin. Extracellular addition of Ca(2+) in the presence of the Ca(2+) ionophore ionomycin produced a sustained elevation in [Ca(2+)](i) that resulted in an inhibition of the cell-to-cell transfer of the fluorescent dye Alexa fluor 594 (IC(50) of 360 nM Ca(2+)). The Ca(2+) dependency of this inhibition of Cx43 gap junctional permeability is very similar to that described in sheep lens epithelial cell cultures that express the three sheep lens connexins (Cx43, Cx44, and Cx49). The intracellular Ca(2+)-mediated decrease in cell-to-cell dye transfer was prevented by an inhibitor of calmodulin action but not by inhibitors of Ca(2+)/calmodulin-dependent protein kinase II or protein kinase C. In experiments that used HeLa cells transfected with a Cx43 COOH-terminus truncation mutant (Cx43(Delta257)), cell-to-cell coupling was similarly decreased by an elevation of [Ca(2+)](i) (IC(50) of 310 nM Ca(2+)) and similarly prevented by the addition of an inhibitor of calmodulin. These data indicate that physiological concentrations of [Ca(2+)](i) regulate the permeability of Cx43 in a calmodulin-dependent manner that does not require the major portion of the COOH terminus of Cx43.     

Oculodentodigital dysplasia-causing connexin43 mutants are non-functional and exhibit dominant effects on wild-type connexin43

Oculodentodigital dysplasia, a rare condition displaying congenital craniofacial deformities and limb abnormalities, has been associated with over 20 known human connexin43 (Cx43) mutations. The localization of two of these mutants, G21R and G138R, was examined in Cx43-positive normal rat kidney cells (NRK) and Cx43-negative gap junctional intercellular communication-deficient HeLa cells. Green fluorescent protein-tagged and untagged Cx43 G21R and G138R mutants were transported to the plasma membrane and formed punctate structures reminiscent of gap junction plaques in both NRK and HeLa cells. Further localization studies revealed no significant trafficking defects as subpopulations of Cx43 mutants were found in both the Golgi apparatus and lysosomes, not unlike wild-type Cx43. Dual patch clamp functional analysis of the mutants expressed in gap junctional intercellular communication-deficient N2A cells revealed that neither G21R nor G138R formed functional gap junction channels, although they successfully reached cell-cell interfaces between cell pairs. Importantly, when either mutant was expressed in NRK cells, dye coupling experiments revealed that both mutants inhibited endogenous Cx43 function. These studies suggest that, although patients suffering from oculodentodigital dysplasia possess one wild-type Cx43 allele, it is likely that Cx43-mediated gap junctional intercellular communication is reduced below 50% because of a dominant-negative effect of mutant Cx43 on wild-type Cx43.     

Cloning and deletion mutagenesis using direct protein-protein interaction on an expression vector. Identification of the calmodulin binding domain of alpha-fodrin

We have screened a lambda gt11 library, constructed with mouse macrophage cDNA, in order to isolate clones that code for calmodulin binding proteins. We have developed a new approach for this purpose using radioactive calmodulin (produced by genetic engineering) to detect fusion proteins that interact with this protein with high affinity. A cDNA clone that codes for mouse macrophage fodrin was isolated, sequenced and identified. By deleting part of the sequence the calmodulin binding domain was located on the fodrin sequence. The site is situated on repeat 11 of fodrin and probably on the extra arm of this repeat. The method we developed is widely applicable to site-directed mutagenesis of interacting proteins.     

Coexpression of connexin 45 with connexin 43 decreases gap junction size

In the human heart, ventricular myocytes express connexin 43 (Cx43) and traces of Cx45. In congestive heart failure, Cx43 levels decrease, Cx45 levels increase and gap junction size decreases. To determine whether alterations of connexin coexpression ratio influence gap junction size, we engineered a rat liver epithelial cell line that endogenously expresses Cx43 to coexpress inducible levels of Cx45 under stimulation of the insect hormone, ponasterone A. In cells induced to express Cx45, gap junction sizes are significantly reduced (by 15% to 20%; p < 0.001), an effect that occurs despite increased levels of junctional connexons made from both connexins. In contrast, coexpression of Cx40 with Cx43 does not lead to any change in gap junction size. These results are consistent with the idea that increased Cx45 expression in the failing ventricle contributes to decreased gap junction size.     

Cardiac mitochondrial connexin 43 regulates apoptosis

Connexin 43 (Cx43) is thought to be present largely in the plasma membrane and its function solely to provide low resistance electrical connection between myocytes. A recent report suggested the presence of Cx43 in the mitochondria as well. We confirmed the presence of Cx43 in the mitochondria isolated from adult rat ventricles with the Cx43 immunoreactivity fractionating to the outer mitochondrial membrane. Mitochondrial Cx43 is mostly phosphorylated only detected by a phospho-specific antibody. Using a Ca2+ -sensitive electrode and Western blot, we showed that the gap junction inhibitors 18-beta-glycyrrhetinic acid (beta-GA), oleamide, and heptanol all induced concomitant release of Ca2+ and cytochrome C in isolated mitochondria whereas the inactive analog 18-beta-glycyrrhizic acid failed to do so. In low density neonatal myocyte culture with no appreciable cell-cell contacts, beta-GA induced apoptosis as assessed by TUNEL staining. Our results suggest a novel role of Cx43 as a regulator of mitochondrial physiology and myocyte apoptosis.     

Differential oligomerization of endoplasmic reticulum-retained connexin43/connexin32 chimeras

To examine early events in connexin oligomerization, we made connexin constructs containing a C-terminal di-lysine based endoplasmic reticulum (ER) retention/retrieval signal (HKKSL). Previously, we found that both Cx32-HKKSL and Cx43-HKKSL were retained in the ER. However, Cx32-HKKSL oligomerized into hexameric hemichannels, but Cx43-HKKSL was retained as an apparent monomer. To define elements that prevent Cx43-HKKSL oligomerization in the ER, we made a series of HKKSL-tagged Cx43/Cx32 chimeras. When expressed by HeLa cells, some chimeras were retained in the ER as apparent monomers, whereas others oligomerized in the ER. To date, the second and third transmembrane domains and the cytoplasmic loop domain provide the minimal sufficient Cx43 element to inhibit ER oligomerization.     

Heteromeric connexin 43/connexin 33 complex endocytosis: A connexin phosphorylation independent mechanism

The role of gap junctions in proliferation, differentiation and apoptosis has been recently highlighted. Nevertheless, the molecular mechanisms that control these physiological events by acting on gap junction channels are still unknown. We have recently demonstrated that heteromeric gap junction plaques composed by Cx43 and Cx33 are unstable at the cell boundary and are rapidly internalized by endocytosis. In the present study, we analyze the phosphorylation status of Cx43 in homomeric (Cx43/Cx43) and heteromeric (Cx33/Cx43) complexes and their association with the tyrosine kinase c-Src. Our data show that c-Src interaction and P2 phosphorylation of Cx43, which are essential for homomeric Cx43 complex endocytosis, were altered in the heteromeric Cx33/Cx43 complex: lack of association between Cx33 and activated c-Src and disappearance of the P2 phosphorylated Cx43 isoform. The present findings demonstrate that the interaction of Cx33 with Cx43 within a same heteromeric complex may conduce to channel instability through alteration of the phosphorylation status of Cx43 independently of the control of the c-Src kinase. The data described here emphasize a new mechanism of Cx43 internalization Src kinase-independent.