Structural insights into the opening mechanism of human Cx43/GJA1 gap junction channel

doi:10.21203/rs.3.rs-5072767/v1

Download PDF

Article

Structural insights into the opening mechanism of human Cx43/GJA1 gap junction channel

https://doi.org/10.21203/rs.3.rs-5072767/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Gating of the gap junction intercellular channel (GJCh) is tightly regulated by several cellular factors; however, the underlying mechanism is poorly understood. A cryo-EM study of human Cx43 GJCh revealed detailed structural changes induced by PIP₂. Cx43 protomers in a phospholipid environment show dynamic equilibrium among several N-terminal helix (NTH) conformations, including gate-covering NTH (GCN) and pore-lining NTH (PLN). Upon treatment with a water-soluble PIP₂ analog, the conformational equilibrium shifted from GCN to PLN in a dose-dependent manner, resulting in a decrease in the pore-occluding density and an increase in the open probability. The PIP₂ head interacts closely with basic residues in the membrane opening between neighboring protomers and the cytoplasmic loop (CL). These ionic interactions strengthen the binding of CL to a transmembrane helix, which consequently inhibits the GCN conformation through steric hindrance. This study provides structural insights into the mechanisms underlying the opening of Cx43 GJCh.

Biological sciences/Structural biology/Electron microscopy/Cryoelectron microscopy

Biological sciences/Molecular biology

connexin 43

GJA1

gap junction

PIP2

cytoplasmic loop

The gap junction intercellular channel (GJCh) is a large pore channel that provides an intercellular passage for ions, metabolites, and signaling molecules¹. A vertebrate GJCh is formed by head-to-head docking of two hemichannels from adjacent cell membranes^{2, 3, 4, 5, 6, 7, 8, 9}, and each hemichannel comprises six connexin protomers. Twenty-one connexin genes were identified and classified into five subfamilies (A–E) based on sequence similarity. Cx43, also known as GJA1, is the most abundant and widely expressed connexin found in humans. It is highly expressed in cardiomyocytes, astrocytes, and osteocytes and plays crucial roles in cardiac contraction¹⁰, neurotransmission¹¹, and mechanotransduction¹². Cx43 GJCh is also important for tumor metabolism and progression¹³, especially in brain cancer¹⁴. Its dysregulation or malfunction causes various human diseases such as oculodentodigital dysplasia (ODDD), erythrokeratodermia variabilis (EKV), and heart diseases^{15, 16, 17, 18}.

Connexins consist of five structural domains: the N-terminal helix (NTH), transmembrane domain (TMD) composed of four transmembrane (TM) helices, extracellular domain (ECD) formed by extracellular loops 1 and 2, cytoplasmic loop (CL), and C-terminal domain (CTD) (Fig. 1b). Connexin family proteins share substantial sequence identities in their NTH, TMD, and ECD regions, and the available structures of Cx26, Cx32, Cx36, Cx43, Cx46, and Cx50 GJChs^{2, 3, 5, 6, 7, 19, 20} show many common structural features in these domains, which may be important for a common functional mechanism of GJChs. However, the highly divergent and relatively flexible CL and CTD have not been structurally determined; thus, the regulation of channel activity through CL and CTD remains unclear.

The gating of Cx43 GJCh is regulated by many factors, such as transjunctional voltage (Vj), intracellular pH, divalent cations, calmodulin, post-translational modifications (PTMs), and membrane lipids^{21, 22, 23, 24, 25, 26, 27}. However, the molecular mechanisms underlying these regulatory processes are not clearly understood. We previously determined the cryo-EM structure of Cx43 GJCh in various lipid/detergent environments and identified three different NTH conformations: gate-covering NTH (GCN), pore-lining NTH (PLN), and flexible-intermediate NTH (FIN)⁴. In the full GCN conformation of Cx43 GJCh, in which all 12 NTHs were in the GCN conformation, the pores were occluded by lipids, suggesting that this structure may represent a closed state. While GCN was strongly induced by high cholesteryl hemisuccinate (CHS) content, removal of CHS from the protein sample largely increased the PLN population and created a dynamic structural equilibrium between the three different connexin conformations. The cryo-EM structures of Cx36 GJCh also show a similar structural equilibrium and pore occlusion by lipids⁵. Based on these data, a lipid-mediated gating mechanism was proposed, which is currently hypothetical because how lipids move in and out of the channel is unclear. In addition, the full PLN conformation of Cx43 GJCh, which is believed to be in the open state, was never the major conformation under all tested conditions, suggesting that the direct binding of an unknown factor may be required for channel opening.

Phosphatidylinositol-4,5-bisphosphate (also known as PIP₂ or PtdIns(4,5)P₂) is mainly localized to the inner leaflet of the plasma membrane and accounts for 1–3% of total lipids. It is an important signaling lipid that regulates various membrane channels²⁸. We also investigated the regulation of Cx43 GJCh by PIP₂. When PIP₂ was depleted in Rat-1 fibroblasts via the translocation of phosphoinositide 5-phosphatase to the plasma membrane, the intercellular transfer of the fluorescent dye Calcein largely decreased²⁹. The overall and local depletion of PIP₂ by wortmannin treatment and Gα_q stimulation in cardiomyocytes, respectively, also reduced the transfer of Lucifer yellow and/or conduction velocity³⁰. Restoration of decreased channel activity was delayed by inhibiting PIP₂ re-synthesis^{29, 30}, confirming that PIP₂ activates Cx43 GJCh. However, it remains unclear how PIP₂ increases channel activity or whether it directly binds to channels.

To understand the opening mechanism of Cx43 GJCh, we performed cryo-electron microscopy (cryo-EM) on Cx43 GJChs in dynamic structural equilibrium. We treated the purified GJCh-nanodisc samples with a soluble PIP₂ analog and analyzed the structures of the individual protomers in the GJCh particles to determine the change in structural equilibrium from GCN to PLN. High-resolution 3D structures reconstructed from GJCh images mainly in the PLN conformation allowed us to reveal the PIP₂-binding site, ordered CL structure, and detailed interactions with PIP₂. Based on these structural data, we propose a lipid-mediated opening mechanism for Cx43 GJCh.

PIP₂ induces the conformational change in Cx43 GJCh from GCN to PLN

In our previous structural study of Cx43 GJCh, we identified one condition (condition 8) in which the ratio of GCN, FIN, and PLN protomers was approximately 2:1:2 (37.3%, 22%, and 40.7%, respectively) (Fig. 1a, b), whereas the protomers in other conditions were mainly in the GCN conformation⁴. Under condition 8, the Cx43-M257 construct with a partial deletion of CTD was used; GJChs were reconstituted into lipid nanodiscs mainly containing 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine (POPE); and the CHS content was largely reduced during nanodisc reconstitution and purification. As Cx43 GJCh was in its most dynamic conformational equilibrium under condition 8, we chose this condition to test the effects of regulatory factors on equilibrium. Notably, aliquots of the GJCh-nanodisc sample in condition 8 were cryo-cooled, stored at − 80 ℃, and used for both previous and current studies. This excludes the possibility of variations that may have occurred during sample preparation.

To examine the effect of PIP₂ on the structural equilibrium of Cx43 GJCh, we used a water-soluble PIP₂ analog, dioctanoylglycerol-PIP₂ (diC8-PIP₂), because PIP₂ cannot be delivered to lipid nanodiscs without detergents. Detergents should be avoided because they affect the structure of GJChs and lipid nanodiscs by binding to GJChs and extracting lipids from the nanodiscs. diC8-PIP₂ has shorter hydrophobic tails but an identical head group compared to PIP₂ and has been effectively used to study the regulation of various ion channels by PIP₂^{28, 31}.

The GJCh-nanodisc sample (5.9 mg/ml) in condition 8 was treated with diC8-PIP₂ at a final concentration of 1.3 mM, which corresponded to an approximately eight-fold molar excess of the Cx43 protomer. After 10 min of incubation, the sample (referred to as the Cx43-PIP₂-8x sample) was loaded onto cryo-EM grids and cryo-cooled. The high protein concentration of 5.9 mg/ml was required to obtain various particle orientations. Several rounds of 2D classification were conducted to remove aggregated, overlapping, or false-positive (such as hemichannel) particles. Three-dimensional classification and refinement produced a reasonably good consensus map at 3.06 Å resolution (Supplementary Figs. 1a, c, 2a). Using 26,825 particle images to reconstitute the consensus map, we performed a protomer-focused 3D classification to determine the ratio of the different NTH conformations (Fig. 1a). Of the resulting five classes, three classes, accounting for 62.3% of the particles, showed the PLN conformation, whereas one class, including 20.4% of the particles, exhibited the GCN conformation (Fig. 1c). This indicates that 1.3 mM diC8-PIP₂ increased the PLN population and decreased the GCN population by approximately 20%.

Because a substantial portion of the protomers were still in the FIN or GCN conformations in the presence of 1.3 mM diC8-PIP₂, we almost doubled the diC8-PIP₂ concentration to produce the Cx43-PIP₂-16x sample and performed a similar cryo-EM experiment to obtain the consensus map at 3.01 Å resolution (Supplementary Figs. 1b, d, 2b). In the protomer-focused 3D classification conducted with 5,720 particle images, we found no protomer classes in the GCN conformation and three classes in the PLN conformation, accounting for 71.6% (Fig. 1d). Therefore, we concluded that diC8-PIP₂ shifted the conformational equilibrium from GCN to PLN in a concentration-dependent manner. However, the percentage of protomers in the FIN conformation (FIN protomers) was not reduced by high diC8-PIP₂ concentration, suggesting that Cx43 GJCh remains in dynamic equilibrium between PLN and FIN, even in the presence of PIP₂.

In the previous structural data without PIP₂, strong pore-occluding densities were observed in a 3D structural class that mainly contained GCN protomers (Supplementary Fig. 3e-g). In contrast, the GCN conformation almost disappeared in the Cx43-PIP₂-16x data, and the C1 consensus map, reconstructed without symmetry imposition, showed much weaker pore-occluding densities (Supplementary Fig. 3a-d). This suggests that pore-occluding lipids diffuse out of the pores during conformational changes from GCN to PLN after treatment with diC8-PIP₂.

Structure of Cx43 GJCh in complex with PIP

Because diC8-PIP₂ increases the PLN protomer population, this small molecule likely binds to and stabilizes the PLN conformation. To identify the map density of diC-PIP₂, we determined the structure of Cx43-PIP₂-16x GJCh in full PLN conformation. Because the Cx43-PIP₂-16x dataset contained approximately 30% FIN protomers, we conducted 3D classification using a mask covering the protein to increase conformational homogeneity. While one class, including 77.4% of the GJCh particles, showed clear PLN density, the other class did not, suggesting that GJChs containing more FIN protomers were effectively separated from those containing more PLN protomers. We selected the former class for the ensuing refinement process and obtained high quality map density at 3.01 Å resolution (Fig. 2a and Supplementary Figs. 1d, 2c, d).

In the refined map of the GJCh nanodisc with D6 symmetry, we identified strong densities with clear features of the PIP head group, including an inositol 1,4,5-trisphosphate and a glycerol backbone at large gaps (membrane openings) between neighboring protomers (Fig. 2a-c). Such densities were not observed at the same position in the previous structure of the PLN conformation under diC8-PIP₂-free conditions, confirming that the densities originated from the diC8-PIP₂ added to the purified GJCh nanodisc sample (Supplementary Fig. 4a, b). Based on the clarity and connectivity of diC8-PIP₂ density, we built a ligand model that included the entire head and parts of the two tails.

Cx43 GJCh in the GCN conformation has no membrane opening between the protomers; thus, PIP₂ does not bind to the channel in this conformation. The conformational change from GCN to PLN is accompanied by the α-π transition in the middle of TM1 that creates the membrane opening. Therefore, PIP₂ binding to the channel likely inhibits the reverse transition (PLN to GCN) by preventing the closure of the membrane opening.

Structure of CL stabilized by PIP binding

We previously found that the structure of Cx43 GJCh in the PLN conformation in the absence of PIP₂ showed a large density blob next to the cytoplasmic protrusion of TM2 (residues E103–K114). However, the density was too low to determine its identity. In the structure of Cx43-PIP₂-16x, the corresponding density was stronger, was a long chain of approximately 25 amino acids, and showed a clear feature of an α-helix extending parallel to the cytoplasmic protrusion of TM2 (Fig. 2a-c). In addition, one end of this density was connected to the N-terminus of TM3, and the other end was located close to the C-terminus of TM2. Because the CL was a 36-amino-acid-long polypeptide connecting TM2 and TM3 and had an α helix region in the secondary structure prediction, we were confident that the density was a part of the CL (Fig. 2d).

The main-chain connectivity of the density was clear enough to build a structural model of the CL, although most of the side chain densities were unclear. Nevertheless, Y137 at the C-terminus of the α-helical region (residues from H126 to Y137; referred to as CL-α) showed a clear side chain density and thus was a landmark for correct amino acid registration (Fig. 2c and Supplementary Fig. 5). The final model exhibited reasonable stereochemistry. For examples, 1) the CL formed an extensive hydrophobic interaction network with TM2 including the interaction of L127, I130, Y137, I139, and V145 in CL with V96, V99, M100, K102, L106, and L113 in TM2 (Fig. 3a, b), 2) K134 in CL-α participated in ionic interactions with E103 and E110 in TM2 (Fig. 3c), and 3) M147 was buried in a hydrophobic pocket formed by T19, L152, Y155, and V231 (Fig. 3d).

We also performed computational prediction of a hexameric Cx43 structure using AlphaFold2^{32, 33} to identify the evolutionarily conserved interactions of CL with TM2. As the CL structures in the five top-ranked models were almost identical, we selected the first model and examined its per-residue confidence score (pLDDT) (Supplementary Fig. 6a-c). Notably, pLDDT values for the residues in CL-α and the cytoplasmic protrusion of TM2 were more than 80, indicating that the interaction between these two helices in the predicted model has high confidence. These regions were superposed onto the corresponding region in the experimental model with the average Cα RMSD of 2.1 Å (Supplementary Fig. 6d), and the amino acid positions in the regions were identical between the two models. These data suggest that the interaction between CL-α and TM2 is evolutionarily important.

The loop structure for the residues G138-G160 in the CL (referred to as CL-β) showed substantial difference between the predicted and experimental models (Supplementary Fig. 6d). The pLDDT values for this region were relatively low, suggesting that the interactions of these residues with other residues in the protein are not evolutionarily conserved (Supplementary Fig. 6b, c). In the experimental model, these two lysine residues interact with the 5' phosphate group on the inositol headgroup of diC8-PIP₂, accounting for how PIP₂-binding contributes to the stabilization of the CL structure (Fig. 4c, d).

We found no interaction of the CL with NTH or the NTH-TM1 loop except a close interaction of M147 with T19 and a distant interaction of V145 with Y17 at the inter-atomic distance of 6.6 Å (Fig. 3e), suggesting that the CL might not contribute much to the formation of PLN. However, the location of CL-β in the PLN conformation was largely overlapped with that of NTH in the GCN conformation, and thus the structured CL could contribute to the maintenance of PLN by inhibiting the formation of GCN (Fig. 3f).

Detailed interaction of Cx43 GJCh with diC8-PIP

In the structure of Cx43-PIP₂-16x GJCh, the PIP₂-binding site is formed by TM1, TM4, and CL-β from a Cx43 protomer and TM2 from the adjacent protomer. The cytoplasmic region of the binding site had highly basic surfaces created by seven lysine residues and one arginine (Fig. 4a). Six lysine residues interact with 4' and 5' phosphate groups on the inositol head, which are the unique feature of PIP₂ distinct from other phospholipids in human cells (Fig. 4c, d). Especially, K146 in CL-β and K105 in TM2, closely interact with 5′ and 4′ phosphates, respectively, within a distance of 3.2 Å. In addition, three amino groups from K102 and K109 in TM2 and K237 in TM4 could be located within approximately 5 Å from 4' phosphate and thus may contribute to the binding affinity for PIP₂. The K102N and K144E mutations cause ODDD^{34, 35}, suggesting that normal channel function might require a strong interaction between Cx43 GJCh and PIP₂.

K109 and K234 are highly conserved basic residues in the human connexin family, and K105, K144, and K146 are moderately conserved (Fig. 4e), suggesting that other connexin GJChs may have similar PIP₂-binding sites. In addition, although the length and amino acid sequence of the CL domain are highly diverse among human connexins, the conservation of K144 and K146 suggests that PIP₂ may induce conformational changes in the CL domains of other connexins to regulate various GJChs.

The 1′ phosphate group on the inositol head interacts closely with R101 and distantly with K23 (Fig. 4c, d). The glycerol backbone and the second carbon atoms in the hydrocarbon tails participated in hydrophobic interactions with A20, Y98, and Y230 (Fig. 4b). As phosphate and glycerol groups are shared by all phosphatidyl lipids, these interactions are not specific to PIP₂. Although R101 is completely conserved only in A-class connexins, the other four residues are highly conserved in the human connexin family: K23 residue as lysine or arginine residues, A20 as hydrophobic amino acids, and Y98 and Y230 as tyrosine or histidine residues (Fig. 4e). Therefore, the interaction of these residues with phosphatidyl lipids at the membrane opening may be important for the function or regulation of connexin GJChs.

It is unclear how GJCh removes pore-occluding lipids to form completely hydrophilic pores without using chemical energy. Our cryo-EM study of Cx43 GJCh elucidated detailed structural changes during channel opening in a well-defined in vitro system, allowing us to propose the following opening mechanism (Fig. 5a, b): 1) PIP₂ in the inner leaflet of the lipid bilayer binds to the cleft between the PLN protomers (membrane opening), mainly through the ionic interaction of the acidic PIP₂ head with the basic cleft. This stabilizes the TMD conformation in the PLN state including the π-helix in the middle of TM1. 2) The interaction of the PIP₂ head with lysine residues in CL-β, which stabilize the CL structure bound to TM2. This inhibits the GCN conformation through steric hindrance and thus stabilizes the PLN conformation. 3) Lipids in the pore lumen diffuse out of the channel through membrane openings that are not occupied by PIP₂.

In a previous structural study on Cx43 GJCh, we proposed a hypothesis of lipid-mediated structural equilibrium based on the observation of various channel conformations and pore-occluding densities in cryo-EM images of GJCh nanodiscs^{4, 5}. However, we cannot exclude the possibility that structural changes occurred during nanodisc reconstruction, during which pore-occluding lipids could be delivered from lipid-detergent micelles. The current study was designed to address this concern using a soluble PIP₂ species and to prove the structural changes in GJCh and the movement of lipids out of the pores in a lipid bilayer environment. Another recent study by our group showed that a soluble gap junction inhibitor, carbenoxolone, induced pore occlusion of Cx36 and Cx43 GJChs by lipids in similar nanodisc systems³⁶. Therefore, the diffusion of lipids, both in and out of the channel, was demonstrated in two studies.

Previous cell-based studies have shown a decrease in intercellular dye transfer activity following artificial PIP₂ depletion in the mammalian cells^{29, 30}. This study proposed a plausible explanation for these observations. The decrease in PIP₂ levels likely caused the dissociation of PIP₂ from Cx43 GJChs, and the consequent conformational change from PLN to GCN induced closure of GJChs. However, as PIP₂ participates in various cellular processes, it remains unclear whether the direct binding of PIP₂ is the main reason for channel regulation. Nevertheless, our in vitro experiments suggested that PIP₂ at physiological concentrations has a substantial and direct effect on regulation for the following reasons: first, while the diC8-PIP₂ concentration of 2.3 mM that induced an almost complete opening of all GJChs corresponds to approximately 0.2% (w/v) in water, cellular PIP₂ accounts for approximately 1% (w/w) of membrane lipids²⁸. Second, because PIP₂ is mostly found in the inner leaflet, its effective concentration is twice as high. Third, a complete shift in the structural equilibrium to PLN may not be necessary for effective regulation, as only a partial shift would substantially increase the number of GJCh in the full PLN conformation.

The regulation of Cx43 GJCh through the CL and CTD has been studied for more than two decades^{37, 38}, but the underlying mechanisms remain unclear. Although the CL-CTD interaction has been thoroughly studied^{38, 39}, the connection between this interaction and channel gating remains unknown. Our structural study, together with previous biochemical studies, provides a plausible explanation for how CL participates in the regulation by pH and phosphorylation. In the wild-type Cx43 GJCh, CTD and TM2 likely compete for interaction with CL, and thus, the partial deletion of CTD would strengthen the TM2-CL interaction, resulting in a shift in the structural equilibrium to the open conformation (PLN). An acidic pH (pH 5.8) was reported to enhance the dimerization of CTD, resulting in an increase in its binding affinity for CL, which might have caused a shift in the structural equilibrium to the closed (GCN) conformation^{39, 40}. Basic pH or phosphorylation of CL or CTD may also strengthen the CL-CTD interaction and weaken the CL-TM2 interaction to downregulate the channel. In our previous structural study on wild-type Cx43 GJCh, we observed an increase in the PLN/GCN ratio according to the deletion of CTD or a pH change from 8.0 to 6.9, which supports these interpretations.⁴

Many missense mutations in CL cause ODDD, but the disease-causing mechanisms are not yet understood^{34, 35}. Many of these mutations, such as I130T, G138D/R/S, G143D/S, K144E, V145G, M147T, and R148G/Q, likely lowered the stability of the CL structure (Fig. 3g). I130, K144, V145, and M147 play important structural roles, and G138 and G143 likely relieve the torsion of the main chain in CL-β. However, because CL interacts with the CTD or Ca²⁺-calmodulin for GJCh regulation, these factors need to be comprehensively considered to understand the disease-causing mechanisms of CL mutations.

Connexin family proteins share high sequence identity in their structured regions, including NTH, TMD, and ECLs, suggesting that they might have common functional mechanisms.^{4, 41} Cx43 and Cx36 GJChs, which belong to the phylogenetically distant A and D classes, respectively, show similar structural changes in NTH and TM1, and pore occlusion by lipids^{4, 5}. Cholesterol derivatives or soluble inhibitors (CBX and mefloquine) commonly induce pore occlusion by lipids in GJChs^{36, 42}. The residues lining the PIP₂-binding site of Cx43 are highly conserved in many other connexins. Taken together, these data suggest that lipid-mediated conformational changes may be a common gating mechanism of vertebrate connexin GJChs. Since pore occlusion by lipids has also been observed in an innexin hemichannel⁴³, a similar gating mechanism may also be conserved in invertebrate GJChs. This possibility needs to be investigated using cryo-EM single-particle analysis of other GJChs. In addition, since PIP₂ participates in various cellular processes such as cytoskeleton dynamics and Ca²⁺ signaling, how the regulation of gap junctions is related to other cellular processes is an important question for future studies.

Protein expression and purification of human Cx43-M257

Cx43-M257 proteins were expressed and purified as described previously⁴. For expression of Cx43-M257, HEK293E cells grown as suspension cells at 37 ℃ were transfected with the vector containing Cx43-M257 at a cell density of ~ 0.6 × 10⁶ cells/ml and was added with dimethyl sulfoxide (DMSO) to a final concentration of 1% to increase transfection efficiency. After transfection, the cells were incubated at 33 ℃, supplemented with tryptone to a final concentration of 0.5% after 48 hours to increase protein expression, and harvested after further incubation for 48 hours.

All purification steps were carried out at 4 ℃ unless indicated otherwise. Harvested cells were lysed using a Dounce homogenizer in buffer A [20 mM Tris (pH 8.0), 250 mM NaCl, 10% glycerol, 0.1 mM phenylmethylsulfonyl fluoride, RNase A, Staphylococcus aureus nuclease, and one tablet of EDTA-free Pierce protease inhibitor tablets]. After centrifugation at 35,000 g for 1 hour, the pellets were washed in buffer B [20 mM (pH 8.0), 1 M NaCl, 1 mM EDTA, 2% glycerol, and protease inhibitor cocktail]. Membrane proteins were extracted with buffer C [20 mM CAPS (pH 10.5), 150 mM KCl, 2% glycerol, 2 mM β-mercaptoethanol, 1 mM EDTA, 0.5/0.05% (w/v) LMNG/CHS, and protease inhibitor cocktail] and incubated for 1 hour with gentle rotation. After centrifugation at 20,000 g for 1 hour, the supernatant was loaded onto a pre-equilibrated rho-1D4 antibody-conjugated agarose resin and incubated for 2 hours with gentle rotation. Agarose resins were washed with 30 column volumes of buffer D [20 mM Tris (pH 8.0), 150 mM KCl, 2% glycerol, 2 mM β-mercaptoethanol, and 0.005/0.0005% (w/v) LMNG/CHS] and bound proteins were eluted by adding excess HRV 3C protease, followed by overnight incubation. The eluted proteins were loaded onto a Superose 6 Increase 10/300 column (GE Healthcare) pre-equilibrated with buffer E [20 mM Tris (pH 8.0), 150 mM KCl, 2 mM β-mercaptoethanol, and 0.005/0.0005% (w/v) LMNG/CHS]. Peak fractions were collected, concentrated, and used for nanodisc reconstitution.

Nanodisc reconstitution.

MSP1E1 proteins were prepared as described previously⁴⁴. Briefly, the proteins were expressed using E. coli BL21(DE3) cells, and purified with Ni-NTA resin and HiPrep 26/10 desalting column (GE Healthcare). The purified MSP1E1 samples were supplemented with 1 mM EDTA, concentrated, and stored at − 80 ℃.

Purified Cx43-M257 proteins were reconstituted into lipid nanodiscs as previously described^{4, 44}. All the reconstitution steps were carried out at 4 ℃. Purified Cx43-M257 proteins were concentrated to 3 mg/ml and mixed with 1-Palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine stock solubilized in 5% (w/v) LMNG at the 1:90 molar ratio of protomer to lipid, followed by incubation for 30 min. The Cx43-lipid mixtures were supplemented with MSP1E1 at the 1:0.3 molar ratio of protomer to MSP1E1 and incubated for 30 min with gentle rotation. For removal of detergent, Bio-Beads SM2 absorbent (Bio-Rad, 100 mg per 1 ml mixture) were incubated for 4 hours, subsequently exchanged with fresh Bio-Beads and incubated overnight. After removal of Bio-Beads, the reconstituted mixtures were loaded onto a Superose 6 Increase 10/300 column pre-equilibrated with buffer F [20 mM Tris (pH 8.0), 150 mM KCl, and 2 mM β-mercaptoethanol]. Peak fractions corresponding to Cx43-M257 samples in nanodiscs were collected, concentrated to 5.9 mg/ml, and used for further cryo-EM experiments. The reconstitution quality was assessed by SDS-PAGE.

Cryo-EM grid preparation and data collection

The reconstituted Cx43-lipid-nanodisc samples (5.9 mg/ml, ~ 193 µM Cx43 protomers) were treated with 10 mM diC8-PIP₂ to the final concentration of 1.3 mM and 2.3 mM (approximately 1:8 and 1:16 molar ratios of Cx43 protomer to diC8-PIP₂) for Cx43-PIP₂-8x and Cx43-PIP₂-16x datasets, respectively. After the addition of diC8-PIP₂, 3 µl of samples were incubated for 10 min at 4 ℃ and loaded onto holey carbon grids glow-discharged under the conditions of 15 mA current, negative charge, and 60 sec. Quantifoil R1.2/1.3 Cu 200 mesh grid and Quantifoil R2/2 Cu 200 mesh grid were used for Cx43-PIP₂-8x and Cx43-PIP₂-16x samples, respectively. The sample-loaded grids were quickly transferred to Vitrobot Mark IV (Thermo Fisher Scientific) for vitrification and blotted at 4 ℃ with 95% humidity for 3 sec and 6 sec for Cx43-PIP₂-8x and Cx43-PIP₂-16x samples, respectively.

The Cx43-PIP₂-8x dataset was collected at Institute for Basic Science (IBS) using a Titan Krios G4 microscope (Thermo Fisher Scientific) operating at 300 kV equipped with a K3 detector (Gatan) and BioQuantum energy filter with a slit width of 20 eV. The dataset was acquired at a nominal magnification of ×105,000 with a pixel size of 0.849 Å/pix and a total accumulated dose of 48.95 e⁻/Å² within 55 fractions, resulting in a dose per fraction of 0.89 e⁻/Å²/F. The Cx43-PIP₂-16x dataset was collected at KAIST Analysis center for Research Advancement (KARA) using a Glacios microscope (Thermo Fisher Scientific) operating at 200 kV equipped with a Falcon 3EC detector (Thermo Fisher Scientific). The dataset was acquired at a nominal magnification of ×120,000 with a pixel size of 0.8943 Å/pix and a total accumulated dose of 40 e⁻/Å² within 40 fractions, resulting in a dose per fraction of 1 e⁻/Å²/F. Detailed data acquisition conditions and parameters are provided in Supplementary Table 1.

Cryo-EM data processing

Cx43-PIP₂-8x and Cx43-PIP₂-16x datasets were processed to reconstruct well-aligned consensus maps in similar procedures using cryoSPARC version 4.2.1 software⁴⁵. Beam-induced motion correction and dose weighting of raw movies were performed using Patch motion correction, followed by contrast transfer function (CTF) estimation using Patch CTF estimation. Unsuitable micrographs were removed semiautomatically and manually based on the parameters such as defocus, astigmatism, CTF fit resolution, and total motion distance. Template-based particle picking using Template Picker was performed; we picked as many particles as possible with a low threshold because the Cx43 grid samples usually shows strong orientation preference. After several rounds of 2D and 3D classifications to remove bad classes, 3D consensus maps were reconstructed as described in detail below.

For the Cx43-PIP₂-8x dataset, 13,918,547 particle images from 11,383 micrographs were picked by Template Picker and down-sampled to 90 × 90 pixel images (3.396 Å/pix) using Fourier cropping. After 9 rounds of 2D classification to remove impurities/artifacts and poorly aligned particles, the remaining particles were subjected to Rebalance 2D Classes to reduce the ‘top view’ proportion (in the preferred orientation) resulting in 83,577 selected particles, which were converted back to 360 × 360 pixel images. Ab-Initio Reconstruction for 3D reference and several rounds of Heterogenous Refinement were performed. One of two 3D classes showing clear secondary structures were selected. The 26,825 particles from the selected classes were subjected to Non-uniform refinement⁴⁶ to reconstruct the final consensus map in ~ 3.1 Å resolution.

For the Cx43-PIP₂-16x dataset, 3,060,486 particle images from 3,089 micrographs were picked by Template Picker and down-sampled to 90 × 90 pixel images (3.577 Å/pix) using Fourier cropping. After 6 rounds of 2D classification, 5,720 selected particles were converted back to 360 × 360 pixel images. Ab-Initio Reconstruction and Homogenous Refinement produced the 3D reconstruction map in 3.1 Å resolution. Beam-induced motion of particles was estimated and corrected by Local Motion Correction. Particle images were subjected to another round of Homogenous Refinement to generate the final consensus map at 3.0 Å resolution, which was used for the protomer-focused conformational variation analysis. Since most GJCh particles in this map consisted of protomers in the PLN conformation, further classification using the final consensus map was performed to reconstruct the PIP₂-bound Cx43 GJCh structure in the full PLN conformation. Particle images used for the final consensus map were further classified by 3D Classification in cryoSPARC version 4.2.1 software without orientation search and with a mask focusing on the protein region excluding micelles. One class including 4,412 particles showed clearer PLN densities. These particles were extracted into 480 × 480 pixel images and subjected to Local Refinement, which generated a 3.0 Å resolution map with D6 symmetry. This map was used for model building.

Map resolutions were estimated based on the 0.143 Fourier Shell Correlation (FSC) criterion⁴⁷. The particle images comprising each consensus map were subjected to protomer-focused conformational variation analyses in RELION 3.1⁴⁸. The detailed procedure is described in our previous report.⁴

Model building and validation.

The previous Cx43 GJCh structure in the PLN conformation (PDB code: 7XQJ) was used as the initial model. A protomer model was fitted to the final cryo-EM map of the Cx43-PIP₂-16x dataset in UCSF Chimera. The CL region (residues from 115 to 150) was manually built in Coot software. The hexameric sub-model was first built and refined using phenix.real_space_refine in PHENIX version 1.21.1 software⁴⁹. The options of Global minimization, secondary structure restraints, non-crystallographic symmetry (NCS) constraints, and rigid-body refinement were applied during the refinement processes. Next, the GJCh model was produced using Apply NCS operators in PHENIX and refined using the same options. To produce the diC8-PIP₂ model, the geometric restraints were generated from canonical SMILE string and optimized using the eLBOW module in PHENIX software.

For validating the structural model, FSC curves were calculated between EM map and model. The model quality was evaluated using MolProbity in PHENIX. Detailed statistics for model refinement and validation are presented in Supplementary Table 1. Figures were made using ChimeraX version 1.7.1 software.

Model prediction

The hemichannel sub-structure of Cx43-M257 was predicted using AlphaFold2-Multimer v2.2.4³³. With the input sequence file in FASTA format containing six Cx43-M257 sequences, we executed the AlphaFold2 multimer run, which generated PKL files containing 25 relaxed models. The pLDDT values of Chain A in the rank 1 model was extracted using a custom Python script. The pLDDT graph was produced using Microsoft Excel.

Data availability

All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Information. The atomic coordinate and cryo-EM density maps have been deposited in PDB and EMDB as follows: Consensus map of Cx43/GJA1 gap junction channel in the presence of diC8-PIP₂ (8-fold molar excess), EMD-38220; Consensus map of Cx43/GJA1 gap junction channel in the presence of diC8-PIP₂ (16-fold molar excess), EMD-38221; Structure of Cx43/GJA1 gap junction intercellular channel in complex with diC8-PIP₂, PDB 8XBM and EMD-38223. The C1 consensus map of Cx43-PIP₂-16x is included in EMD-38221 as an additional volume.

Acknowledgements

We thank Bumhan Ryu at IBS, Hyeongseop Jeong at KBSI, and Jin-Seok Choi at KARA for their assistance in data collection. This work was supported by the Suh Kyungbae Foundation (SUHF-18010097 to J.-S.W.), the National Research Foundation (NRF to J.-S.W.) grants funded by the Ministry of Science and ICT (RS-2023-00217798 to J.-S.W.), and a Korea University Grant.

Author contributions

J.-S.W. conceived this project; H.-J.L. and H.J.C. purified GJChs; H.-J.L. performed electron microscopy; H.-J.L. and J.-S.O analyzed the EM data and determined the structures; H.-J.L., J.-S.O, H.J.C., and J.-S.W. wrote the manuscript.

Competing interests

The authors declare no competing interests.

Goodenough DA, Paul DL (2009) Gap junctions. Cold Spring Harb Perspect Biol 1:a002576
Bennett BC et al (2016) An electrostatic mechanism for Ca(2+)-mediated regulation of gap junction channels. Nat Commun 7:8770
Flores JA et al (2020) Connexin-46/50 in a dynamic lipid environment resolved by CryoEM at 1.9 A. Nat Commun 11:4331
Lee HJ et al (2023) Conformational changes in the human Cx43/GJA1 gap junction channel visualized using cryo-EM. Nat Commun 14:931
Lee SN et al (2023) Cryo-EM structures of human Cx36/GJD2 neuronal gap junction channel. Nat Commun 14:1347
Maeda S et al (2009) Structure of the connexin 26 gap junction channel at 3.5 A resolution. Nature 458:597–602
Myers JB et al (2018) Structure of native lens connexin 46/50 intercellular channels by cryo-EM. Nature 564:372–377
Unger VM, Kumar NM, Gilula NB, Yeager M (1999) Three-dimensional structure of a recombinant gap junction membrane channel. Science 283:1176–1180
Unwin PN, Zampighi G (1980) Structure of the junction between communicating cells. Nature 283:545–549
Rodriguez-Sinovas A, Sanchez JA, Valls-Lacalle L, Consegal M, Ferreira-Gonzalez I (2021) Connexins in the Heart: Regulation, Function and Involvement in Cardiac Disease. Int J Mol Sci 22
Giaume C, Fromaget C, el Aoumari A, Cordier J, Glowinski J, Gros D (1991) Gap junctions in cultured astrocytes: single-channel currents and characterization of channel-forming protein. Neuron 6:133–143
Plotkin LI, Speacht TL, Donahue HJ (2015) Cx43 and mechanotransduction in bone. Curr Osteoporos Rep 13:67–72
Zhou M et al (2023) The roles of connexins and gap junctions in the progression of cancer. Cell Commun Signal 21:8
Chen Q et al (2016) Carcinoma-astrocyte gap junctions promote brain metastasis by cGAMP transfer. Nature 533:493–498
Srinivas M, Verselis VK, White TW (2018) Human diseases associated with connexin mutations. Biochim Biophys Acta Biomembr 1860:192–201
Boyden LM et al (2015) Dominant De Novo Mutations in GJA1 Cause Erythrokeratodermia Variabilis et Progressiva, without Features of Oculodentodigital Dysplasia. J Invest Dermatol 135:1540–1547
Michela P, Velia V, Aldo P, Ada P (2015) Role of connexin 43 in cardiovascular diseases. Eur J Pharmacol 768:71–76
Bonacquisti EE, Nguyen J, Connexin (2019) (Cx43) in cancer: Implications for therapeutic approaches via gap junctions. Cancer Lett 442 43:439–444
Qi C et al (2023) Structure of the connexin-43 gap junction channel in a putative closed state. Elife 12
Qi C et al (2023) Structures of wild-type and selected CMT1X mutant connexin 32 gap junction channels and hemichannels. Sci Adv 9:eadh4890
Bukauskas FF, Bukauskiene A, Verselis VK (2002) Conductance and permeability of the residual state of connexin43 gap junction channels. J Gen Physiol 119:171–185
Peracchia C (2004) Chemical gating of gap junction channels; roles of calcium, pH and calmodulin. Biochim Biophys Acta 1662:61–80
Bukauskas FF, Verselis VK (2004) Gap junction channel gating. Biochim Biophys Acta 1662:42–60
Zou J et al (2014) Cholesterol modulates function of connexin 43 gap junction channel via PKC pathway in H9c2 cells. Biochim Biophys Acta 1838:2019–2025
Malewicz B, Kumar VV, Johnson RG, Baumann WJ (1990) Lipids in gap junction assembly and function. Lipids 25:419–427
Puebla C, Retamal MA, Acuna R, Saez JC (2017) Regulation of Connexin-Based Channels by Fatty Acids. Front Physiol 8:11
Locke D, Harris AL (2009) Connexin channels and phospholipids: association and modulation. BMC Biol 7:52
Suh BC, Hille B (2008) PIP2 is a necessary cofactor for ion channel function: how and why? Annu Rev Biophys 37:175–195
van Zeijl L et al (2007) Regulation of connexin43 gap junctional communication by phosphatidylinositol 4,5-bisphosphate. J Cell Biol 177:881–891
Hofgaard JP et al (2008) Phosphatidylinositol-bisphosphate regulates intercellular coupling in cardiac myocytes. Pflugers Arch 457:303–313
Rohacs T (2014) Phosphoinositide regulation of TRP channels. Handb Exp Pharmacol 223:1143–1176
Jumper J et al (2021) Highly accurate protein structure prediction with AlphaFold. Nature 596:583–589
Evans R et al (2022) Protein complex prediction with AlphaFold-Multimer. bioRxiv, 2021.2010.2004.463034
Paznekas WA et al (2009) GJA1 mutations, variants, and connexin 43 dysfunction as it relates to the oculodentodigital dysplasia phenotype. Hum Mutat 30:724–733
Zheng L et al (2020) Calmodulin Directly Interacts with the Cx43 Carboxyl-Terminus and Cytoplasmic Loop Containing Three ODDD-Linked Mutants (M147T, R148Q, and T154A) that Retain alpha-Helical Structure, but Exhibit Loss-of-Function and Cellular Trafficking Defects. Biomolecules 10
Lee CWJ, Woo HS (2024) J. S. Structural insights into the closing mechanism of gap junction intercellular channels by carbenoxolone.)
Lampe PD, TenBroek EM, Burt JM, Kurata WE, Johnson RG, Lau AF (2000) Phosphorylation of connexin43 on serine368 by protein kinase C regulates gap junctional communication. J Cell Biol 149:1503–1512
Duffy HS et al (2002) pH-dependent intramolecular binding and structure involving Cx43 cytoplasmic domains. J Biol Chem 277:36706–36714
Hirst-Jensen BJ, Sahoo P, Kieken F, Delmar M, Sorgen PL (2007) Characterization of the pH-dependent interaction between the gap junction protein connexin43 carboxyl terminus and cytoplasmic loop domains. J Biol Chem 282:5801–5813
Sorgen PL, Duffy HS, Spray DC, Delmar M (2004) pH-dependent dimerization of the carboxyl terminal domain of Cx43. Biophys J 87:574–581
Lee HJ et al (2020) Cryo-EM structure of human Cx31.3/GJC3 connexin hemichannel. Sci Adv 6:eaba4996
Cho H-J, Lee HH (2023) Mefloquine-induced conformational shift in Cx36 N-terminal helix leading to channel closure mediated by lipid bilayer. bioRxiv, 2023.2012.2022.573018
Burendei B et al (2020) Cryo-EM structures of undocked innexin-6 hemichannels in phospholipids. Sci Adv 6:eaax3157
Ritchie TK et al (2009) Chapter 11 - Reconstitution of membrane proteins in phospholipid bilayer nanodiscs. Methods Enzymol 464:211–231
Punjani A, Rubinstein JL, Fleet DJ, Brubaker MA (2017) cryoSPARC: algorithms for rapid unsupervised cryo-EM structure determination. Nat Methods 14:290–296
Punjani A, Zhang H, Fleet DJ (2020) Non-uniform refinement: adaptive regularization improves single-particle cryo-EM reconstruction. Nat Methods 17:1214–1221
Scheres SH, Chen S (2012) Prevention of overfitting in cryo-EM structure determination. Nat Methods 9:853–854
Zivanov J et al (2018) New tools for automated high-resolution cryo-EM structure determination in RELION-3. Elife 7:e42166
Afonine PV et al (2018) Real-space refinement in PHENIX for cryo-EM and crystallography. Acta Crystallogr D Struct Biol 74:531–544

There is NO Competing Interest.

Download PDF

Version 1

posted

You are reading this latest preprint version

Structural insights into the opening mechanism of human Cx43/GJA1 gap junction channel

Status:

Version 1

Abstract

Figures

Introduction

Results

PIP₂ induces the conformational change in Cx43 GJCh from GCN to PLN

Structure of Cx43 GJCh in complex with PIP

Structure of CL stabilized by PIP binding

Detailed interaction of Cx43 GJCh with diC8-PIP

Discussion

Materials and Methods

Protein expression and purification of human Cx43-M257

Cryo-EM grid preparation and data collection

Cryo-EM data processing

Model prediction

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1

Structural insights into the opening mechanism of human Cx43/GJA1 gap junction channel

Status:

Version 1

Abstract

Figures

Introduction

Results

PIP2 induces the conformational change in Cx43 GJCh from GCN to PLN

Structure of Cx43 GJCh in complex with PIP

Structure of CL stabilized by PIP binding

Detailed interaction of Cx43 GJCh with diC8-PIP

Discussion

Materials and Methods

Protein expression and purification of human Cx43-M257

Cryo-EM grid preparation and data collection

Cryo-EM data processing

Model prediction

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1

PIP₂ induces the conformational change in Cx43 GJCh from GCN to PLN