Abstract
Connexins are ubiquitous channel forming proteins that assemble as plasma membrane hemichannels and as intercellular gap junction channels that directly connect cells. In the heart, gap junction channels electrically connect myocytes and specialized conductive tissues to coordinate the atrial and ventricular contraction/relaxation cycles and pump function. In blood vessels, these channels facilitate long-distance endothelial cell communication, synchronize smooth muscle cell contraction, and support endothelial-smooth muscle cell communication. In the central nervous system they form cellular syncytia and coordinate neural function. Gap junction channels are normally open and hemichannels are normally closed, but pathologic conditions may restrict gap junction communication and promote hemichannel opening, thereby disturbing a delicate cellular communication balance. Until recently, most connexin-targeting agents exhibited little specificity and several off-target effects. Recent work with peptide-based approaches has demonstrated improved specificity and opened avenues for a more rational approach toward independently modulating the function of gap junctions and hemichannels. We here review the role of connexins and their channels in cardiovascular and neurovascular health and disease, focusing on crucial regulatory aspects and identification of potential targets to modify their function. We conclude that peptide-based investigations have raised several new opportunities for interfering with connexins and their channels that may soon allow preservation of gap junction communication, inhibition of hemichannel opening, and mitigation of inflammatory signaling.
I. Introduction
Connexins are ubiquitous integral membrane proteins present in almost all cells of the body. They are strongly expressed in major organs such as the heart, brain, and liver, as well as in endothelial and smooth muscle cells of blood vessels. Their main function is to facilitate cell-cell communication and they do so in the most direct way possible, by forming channels called gap junctions (GJs) that connect the cytoplasm of cells. This short route connection serves as a powerful coordinator of cell function in complex tissues like heart and brain; it also permits efficient long-distance communication along rows of GJ-connected cells, as e.g., in the His-Purkinje conduction system in the heart or in endothelial cells of the blood vessel wall to transmit upstream vasodilatory messages (de Wit and Griffith, 2010). In electrically excitable cells like cardiac myocytes, GJ channels facilitate electrical coupling by allowing cell-to-cell passage of ions. Action potentials spread from one cell to another via GJs that are mainly localized at the cell poles in the plicate and interplicate regions of the intercalated disk (ID) (Spach and Heidlage, 1992). Isolated individual cardiomyocytes do not communicate, but when manipulated into close contact with each other, they start to communicate electrically within a couple of minutes via newly established GJ channels (Weingart and Maurer, 1988). The importance of connexins is clear from mouse knockout studies of the major cardiovascular connexins, which yield a nonviable phenotype for Cx26−/−, Cx37−/− ̶ Cx40−/− double knockouts (KOs), Cx43−/−, and Cx45−/− (reviewed in Simon et al., 1998; Söhl and Willecke, 2004). An example illustrating the importance of connexins in the human body concerns inherited mutations in the GJB2 gene that codes for Cx26, which cause congenital sensorineural deafness that has a prevalence estimated in the order of 1:5000 births (Chan and Chang, 2014; Esseltine and Laird, 2016). Other examples include polymorphisms of GJA4 (Cx37), which are linked to vascular disease, mutations of GJA5 (Cx40), which are known to predispose for atrial fibrillation and GJA1 mutations (Cx43), which are generally not associated with a cardiac phenotype (Pfenniger et al., 2011; Delmar and Makita, 2012; Molica et al., 2014) but may lead to oculodentodigital dysplasia (ODDD), a rare primarily autosomal dominant clinical syndrome characterized by multiple malformations. An overview of connexin genes and chromosome locations can be found in Table 1 of Söhl and Willecke (2004); for the distribution of the various connexins in organs and tissues see Table 2 in Laird (2006).
GJs were discovered half a century ago (Revel and Karnovsky, 1967; Brightman and Reese, 1969), and their connexin building blocks were discovered more than 40 years ago (Goodenough, 1974). GJs are dodecameric channels formed by the interaction of two opposed hexameric hemichannels (HCs), also called connexons. Molecular cloning studies have established that connexins form a family of related proteins (Beyer et al., 1990). Twenty-one connexin genes have been identified in the human genome and 20 in the murine genome, which encode proteins with a molecular mass (MM) that ranges from 23 to 62 kDa (Söhl and Willecke, 2004; Beyer and Berthoud, 2009). Connexins are named according to their MM; they have a tetraspan membrane topology, with four transmembrane (TM) domains, two extracellular loops (EL1, EL2), a cytoplasmic loop (CL), and their N- and C-terminal tails (NT and CT) located inside the cell (Fig. 1). The channels formed by the different connexins often also differ in their gating properties, conductances, and permeabilities to various ions and molecules. In general, GJ channels have a pore diameter in the 10–20 Å range and grant passage not only to atomic ions such as K+, Na+, or Ca2+, but also to metabolic molecules with a MM below ∼1.5 kDa (assuming an approximate spherical shape) like ATP, glucose, ascorbic acid, or glutathione, and second messengers such as cAMP, cGMP, or inositol trisphosphate (IP3) (Alexander and Goldberg, 2003; Saez et al., 2003; Li et al., 2012a). Recent evidence indicates that siRNA and miRNA may also pass through GJs, most likely as a rod-shaped molecule linearly permeating the pore with its smallest dimension (Brink et al., 2012). As a result of the passage of multiple substances, which can exert beneficial but also toxic effects, GJs may also be involved in communicating and spreading harmful messages. Two marked examples of such inadequate signaling include the propagation of inflammation along the blood vessel wall (Parthasarathi et al., 2006) and bystander cell death propagation in ischemic cardiomyocytes and brain cells (Contreras et al., 2004; García-Dorado et al., 2004; Decrock et al., 2009b). GJs can close and uncouple cells in response to various conditions, either as a result of mutations or activation of signaling cascades. As illustrated above, several connexin mutations with impact on the cardiovascular system or the brain have been characterized (reviewed in Pfenniger et al., 2011; Abrams and Scherer, 2012; Delmar and Makita, 2012; De Bock et al., 2013a; Molica et al., 2014). GJs also typically close under ischemic conditions, as a result of intracellular acidification and increased cytoplasmic Ca2+ as forefront signals (Kleber, 1992; Anderson et al., 2003; Dhein, 2006a; Evans, 2015; Moore and O’Brien, 2015). Additionally, GJs are modulated in various ways by posttranslational modifications. In principle, GJs can be targeted by pharmacological inhibitors to counteract their contribution to bystander cell death, but in heart as well as in brain, the success of such an approach is not guaranteed because it will compromise the physiologic roles of GJs, i.e., impulse propagation in the heart or substrate delivery through the GJ-connected astrocytic network (Rouach et al., 2008), which also contributes, albeit partially, to spatially buffering K+ away from zones of high neuronal activity (Wallraff et al., 2006). To preserve physiologic functions, GJs may be targeted to correct or prevent uncoupling. In stroke and spinal cord injury, knockout of specific connexins often gives different outcomes compared with pharmacological GJ inhibition, which may result from channel-independent connexin functions that remain operational in the latter case while absent in knockout. Another player that has entered the field and attracted considerable interest as a novel pharmacological target are the connexin HCs.
Topology of human Cx26 and Cx43 indicating crucial domains as well as peptides that affect protein and channel functions. (Left) Illustrates the extracellular loops (EL1, EL2) of Cx26 (intracellular protein parts not shown), indicating the position of highly conserved Cys residues (three per loop, yellow-filled circles) that act to stabilize the loops. Crucial residues involved in interactions between the loops upon docking of two opposed hemichannels are also indicated; for EL1, Asn-54 (green-filled circle) forms a hydrogen bond with Leu-56 of EL1 from the opposite connexin (orange), whereas Gln-57 (green) forms a hydrogen bond with Gln-57 from the opposite connexin. For EL2, salt bridges are formed between Lys-168, Asp-179, and Thr-177 from one side (green) with Asn-176 from the opposite connexin (orange). (Right) Illustrates the Cx43 topology indicating the location of conserved EL1/EL2 Cys residues (yellow). Several important domains are illustrated including the VCYD and FPISH motifs on EL1, SRPTEK on EL2 and the L2, Gap19, and CaM (calmodulin interaction site) sequences on the cytoplasmic loop (CL). Domains on the C-terminal tail (CT) include the tubulin-binding JM domain (juxtamembrane, crucial for microtubule binding), the drebrin-binding domain (Drb; crucial domain indicated; links Cx43 to F-actin), the Nedd4 domain (ubiquitin ligase), SH3 domain, and the ZO-1─binding domain (links Cx43 to F-actin). Peptides mimicking some of these domains are illustrated in the color of the corresponding domain. RRNY peptide (RRNYRRNY) is not a mimetic peptide; like L2 and Gap19, it interacts with the Cx43 CT and prevents GJ closure while inhibiting HC opening. JM2 peptide is a mouse version that differs in residue 243 from the human. The two yellow-marked CT sequences (315–326 and 340–348) are α-helical domains, whereas the rest of the CT is intrinsically disordered. The light green-marked domains are important Tyr-based sorting sequences involved in Cx43/GJ internalization (Y265AYF, Y286KLV). Blue-filled circles with white amino acid letter codes indicate mutations characterized by increased HC function relative to the function of GJs. Black-filled circles with white letter codes indicate ODDD mutations with reduced GJ and HC function. Red-filled circles with black Tyr or Ser indicate major phosphorylation sites, including CK1-targeted Ser-325,328,330, MAPK-targeted Ser-255,262,279,282, PKC-targeted Ser-368, Akt/PKB-targeted Ser-369,373, and Src/Tyk2-targeted Tyr-247,265,313. Ser-364,365 is phosphorylated by PKA but in an indirect manner involving other kinases. A detailed account on the role of the various amino acids and domains illustrated here can be found in sections II and III.
Although GJs have been investigated for over half a century, the interest in possible functions of HCs has a more recent origin. HCs were supposed to be closed until they interact and form a GJ channel at which point they open. It is only since the early 1990s that it was realized that HCs could open without necessarily forming a GJ, resulting in a conduit that communicates with the extracellular space, not with a neighbor cell (Paul et al., 1991). It is now clear that the spectrum of connexin functions encompasses actions of the connexins themselves (channel-independent functions), functions related to GJs, and functions related to HC opening. GJs have well established physiologic roles but they may additionally exert pathologic effects, e.g., by contributing to bystander cell death (Decrock et al., 2009b, 2017). By contrast, HCs have mainly been implicated in pathologic contexts, although it is not clear whether they have any physiologic function. HCs have roughly the same upper limit of ∼1.5 kDa for passing substances but passage is predicted to be easier as the channel is half as long as a GJ channel (the conductance is twice as large for a HC compared with a GJ channel). HCs allow the free passage of Na+ and K+, which may lead to cell swelling, and facilitate Ca2+ entry (Schalper et al., 2010; Fiori et al., 2012), which may cause cellular Ca2+ overload and promote the escape of various molecules (Chandrasekhar and Bera, 2012). These include ATP (Eltzschig et al., 2006; Kang et al., 2008; Bol et al., 2016), glutamate (Ye et al., 2003), lactate (Karagiannis et al., 2016), NAD+ (Bruzzone et al., 2001), IP3 (Gossman and Zhao, 2008), PGE2 (Cherian et al., 2005; Siller-Jackson et al., 2008; Burra and Jiang, 2009), and glutathione (Rana and Dringen, 2007; Ye et al., 2015) that all potentially may become depleted when the cell is under stress and energy reserves run on empty. All released substances may additionally act as autocrine and paracrine signaling molecules, which has been best documented for ATP (Lohman et al., 2012; Wang et al., 2013b; Lohman and Isakson, 2014). HCs also facilitate molecular entry provided there is a chemical or electrochemical driving force; for example, the fluorescent glucose analog 2-(N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino)-2-deoxyglucose may enter cells through HCs. Based on data from the Human Metabolome Database, Esseltine and Laird (2016) estimated the number of molecules that may be able to escape through open HCs in the order of several tens of thousands (Esseltine and Laird, 2016), underscoring the potential scope of inappropriate HC opening. The mechanisms that lead to HC opening link to mutational defects that result in constitutively open HCs or to triggers of electrical, chemical, or mechanical nature. Mutations may impact various levels of the connexin life cycle, including connexin transport to the plasma membrane, functions of GJs, or HC functions. Interestingly, some mutations result in a gain-of-function of HCs and a loss-of-function of GJs demonstrating that HCs/GJs, although composed of the same building blocks, can in principle be distinctly modulated. Evidence for such distinct regulation is accumulating and further expanded in this review. Some mutations of Cx43 that result in increased HC function, yielding leaky HCs, are illustrated in Fig. 1 (for review, see Retamal et al., 2015). In addition to constitutive mutational defects, HCs can be triggered to open in response to electrical signals, i.e., changes of the membrane potential, mechanical forces acting on the plasma membrane (Siller-Jackson et al., 2008; Batra et al., 2014), and chemical signals including the lowering of extracellular Ca2+, increases of intracellular (cytoplasmic) Ca2+, NO, or more complex chemical environments such as proinflammatory conditions or ischemia-reperfusion (reviewed in John et al., 2003; Wang et al., 2013a; Castellano and Eugenin, 2014; Orellana et al., 2014, Schulz et al., 2015; Orellana, 2016).
Pannexins are another protein family composed of three members (Panx1, Panx2, and Panx3) that form channels that resemble connexin hemichannels (Bruzzone et al., 2003). They share the same tertraspan topology, with two ELs, one CL, and N- and C-terminals inside the cell, but lack sequence homology. They are vertebrate analogs of invertebrate innexins that, in contrast to innexins and connexins, generally do not form cell-cell connecting channels as GJs do. As a result, the consensus is to call hexameric channels composed of Panx1, the most common pannexin, just channels and not HCs (Sosinsky et al., 2011). Pannexins have in common with connexin HCs a wide pore, and both act as forefront diffusive ATP release pathways [estimated pore size is 17–21 Å for Panx1, slightly higher than for connexins, and ∼30 Å for Panx2, which possibly forms octameric channels (Ambrosi et al., 2010)]. Panx1 has been demonstrated to be important in inflammasome activation (Kanneganti et al., 2007), and Panx1 and Panx2 are reported to be involved in ischemia-induced neuronal cell death [Panx1 (MacVicar and Thompson, 2010; Thompson, 2015; Weilinger et al., 2016); Panx2 (Bargiotas et al., 2011)]. Panx1-triggered ATP release acts as a find-me signal of apoptotic cells, which attracts phagocytic cells to remove dying cells early in the apoptotic process in an orderly manner (Chekeni et al., 2010). In the heart, Panx1 ATP release has been implicated in attracting phagocytes (Oishi et al., 2012) in pathologic fibrosis (Lu et al., 2012a), in activating sympathetic fibers (Dong et al., 2016), in ischemic pre- and postconditioning (Vessey et al., 2011a,b), in atrial fibrillation (Petric et al., 2016), and as a large conductance channel in cultured atrial cardiomyocytes (Kienitz et al., 2011) (roles of cardiac pannexins are reviewed in Li et al., 2015). In blood vessels, Panx1 in smooth muscle cells is involved in the regulation of vascular tone (Billaud et al., 2012), whereas its presence in venous endothelial cells regulates leukocyte migration during inflammation (Lohman et al., 2015; reviewed in Good et al., 2015; Begandt et al., 2017).
Panx1 channels can be opened by caspases-3 and -7, which cleave off part of the CT, and by Src family kinases that target the CT (Chekeni et al., 2010; Thompson, 2015; Weilinger et al., 2016). Most of the connexin channel inhibitors also inhibit pannexin channels (Dahl et al., 2013), but Panx1 is specifically inhibited by 10Panx1, a peptide mimicking a sequence on EL1 of Panx1 (Pelegrin and Surprenant, 2006) and the quinolone antibiotic trovafloxacin, which explains the toxicity and side effects of this drug (Poon et al., 2014). Due to the lack of sequence homology with the connexins, the regulation of Panx1 channels is very different from connexins. For example, Panx1 channel open with truncation of the CT (Chekeni et al., 2010), whereas the CT is essential for Cx43 HC function (De Vuyst et al., 2007; Kang et al., 2008; Ponsaerts et al., 2010) (Fig. 2A). In terms of regulation by posttranslational modifications, pannexins still need to be scrutinized in detail. Most importantly, pannexins function as a plasma membrane channel without necessarily connecting cells, whereas connexins represent the more complicated case because they form junctional channels as well as HCs. For these reasons, in this review we chose to concentrate on connexins and limit pannexin discussion to introductory concepts.
Functional effects of loop-tail interactions and [Ca2+]i elevation on Cx43 channel function. (A) Interaction of the connexin C-terminal tail (CT) with the cytoplasmic loop (CL) distinctly influences the function of gap junctions (GJs) and hemichannels (HCs). In the absence of CT-CL interaction, GJs are open while HCs are kept closed. Upon interaction of the CT with the CL, GJs are closed while HCs become available to open. The actual opening of HCs only occurs when a trigger is present, which can be of electrical (changes in membrane potential leading to depolarization or positive voltages) or chemical nature (e.g., changes in extracellular or intracellular Ca2+ concentration, inflammatory conditions, ischemic conditions including reperfusion). HC blockers like L2, Gap19 and RRNY (sequences see Fig. 1) bind to the CT and prevent CT-CL interaction, thereby driving HCs from the available to open state to the closed state. At the level of GJs, this prevents the closure of the junctional channels. (B) [Ca2+]i modulation of Cx43 HC opening. Moderate [Ca2+]i promotes HC opening via calmodulin-dependent signaling (red part of the bell-shaped curve). High [Ca2+]i inhibits HC opening by disrupting CT-CL interaction (blue part). CT9 peptide removes the high [Ca2+]i brake. For details, see sections II and III.
This review starts with an introduction on connexins, their expression, life cycle, and regulation (section II), followed by a more detailed account on recent insights and substances that modulate connexin channel function (section III). We next review the roles of connexins in vascular disease, including atherosclerosis, thrombosis, restenosis, and ischemia-reperfusion injury (section IV); in cardiac disease discussing risk factors, ischemia-reperfusion injury, and cardioprotection (section V); and atrial and ventricular fibrillation (section VI). Section VII gives a detailed account on GJA1 mutations that lead to the clinical syndrome of ODDD. Although a rare disease, it is a resource of mutations that affect Cx43 trafficking as well as channel functions, providing crucial insights into the role of specific Cx43 amino acid positions and domains, in some cases (e.g., I130T) leading to arrhythmogenesis. In the last chapter, we discuss the role of connexins in brain and spinal cord injury, stroke, ocular disease, and neurodegenerative brain disease, which is often associated with vascular alterations (section VIII).
Despite the various roles of connexins and their channels in physiology and pathology, the research field is handicapped because all pharmacological agents (inhibitors or promoters) to date have little specificity and often multiple side effects. However, work in the past two decades demonstrates accumulating evidence that peptide-based molecules are the better option in terms of specificity and off-target effects. Peptide studies are the perfect start and opportunity to define a pharmacophore when combined with systematic amino acid substitution approaches and with surface plasmon resonance (or its more recent congener microscale thermophoresis) and NMR studies, collectively allowing for the characterization of crucial peptide-protein interaction domains. Exemplary for such approach are the L2 and AAP peptides that counteract GJ closure, from which several interesting peptidomimetics have been developed (see section III). Another example is the CT9 peptide, which promotes the formation of GJs by affecting the Cx43 life cycle, and also affects HC function (see sections II and III). A second complication in the connexin field is the fact that two types of channels are formed, GJs and HCs, the functions of which are difficult to disentangle, especially in vivo. Recently, several peptides are in the limelight, including L2 and Gap19 peptides that block Cx43 HCs without inhibiting GJs, and Peptide5 that inhibits HCs but not GJs when applied at low concentration (see section III). Thus, there is new excitement and expectation in the field that connexin-targeting drugs may at some point join the pharmacological arsenal available for therapeutic applications (Naus and Giaume, 2016).
II. Expression, Life Cycle, and Regulation of Connexins
A. Expression of Connexins
The major, most prominently expressed cardiovascular connexins are Cx37 [GJA4 gene on chromosome 1 in the human (Söhl and Willecke, 2003)], Cx40 (GJA5 on chromosome 1), Cx43 (GJA1 on chromosome 6), and Cx45 (GJA7 on chromosome 17), with Cx43 being the most abundant isoform. Connexins have been grouped in five families based on their sequence (conserved domains), length of the cytoplasmic loop, and gene structure, each represented by a Greek letter (Cruciani and Mikalsen, 2006; Abascal and Zardoya, 2013). Cx37, Cx40, and Cx43 belong to group α, Cx32 is a β connexin, and Cx45 is a γ connexin. Their regional expression palette differs between the vascular system and the heart. In both, the expression pattern can vary with the species and we mainly focus on the human in what follows.
Blood vessel endothelial cells, including those of the coronary system that perfuses the heart, express Cx37, Cx40, Cx43, and Cx45 but the level or even presence depends on the diameter of the vessels. Cx37 and Cx43 are predominant at all levels up to capillaries, Cx40 is not present in capillary endothelial cells, and Cx45 is only present in large arteries. On the venous side, the connexin expression is low, and another connexin not mentioned yet, Cx32, is found in veins (for reviews, see Haefliger et al., 2000; Begandt et al., 2017). The smooth muscle cells surrounding the endothelium express Cx37, Cx43, and Cx45; Cx40 expression is low or not present.
In heart muscle, five different connexins are represented: Cx31.9, Cx37, Cx40, Cx43, and Cx45. Cx43 is present primarily in working myocytes of the atrium and ventricle (Davis et al., 1994,1995); it is the predominant connexin in ventricular myocardium (Severs et al., 2008). It is also present in the human conduction system but less so in mice and rat (Teunissen and Bierhuizen, 2004). Cx43 also abounds in fibroblasts, which provide the structural skeleton of the myocardium that strongly expands in heart disease (McArthur et al., 2015); GJs may form between cardiomyocytes and fibroblasts, thereby increasing the arrhythmia risk. Cx43 is also found in stem cells used for cardiotherapeutic purposes (Lu et al., 2012b); as such, they may aid in electrically reconnecting cardiomyocytes and promote conduction (Hofshi et al., 2011). Cx40 is present in the atrium and conducting system (except in rats where Cx43 is the main atrial connexin) (Davis et al., 1994; Saffitz et al., 1994; Simon et al., 1998; van Rijen et al., 2001). Cx40 is prominently expressed in the ventricle early in development, but then declines to near absence in the adult (Van Kempen et al., 1996). Levels of Cx40 and Cx43 are very similar in atrial myocytes (Lin et al., 2010). Cx45 is found in many cardiac regions (Davis et al., 1994), but it may not be abundant; it may be important in defining the developing conduction system (Coppen et al., 1999). Cx31.9 (Cx30.2 in rodents) is a connexin that was discovered relatively recently (Belluardo et al., 2001; Nielsen et al., 2002), which may contribute to atrioventricular nodal impulse conduction (Kreuzberg et al., 2005, 2006) (human Cx31.9 gene is on chromosome 17).
Last, but not least, several types of blood cells also express connexins. Red blood cells do not express connexins but they do express Panx1 (reviewed in Begandt et al., 2017). Platelets express Cx37 and Cx40, with low levels of Cx32 and Cx43 and Panx1 (reviewed in Vaiyapuri et al., 2015; Molica et al., 2017). Cx43 and Cx37 are present in monocytes, and Cx43 is present in neutrophils and T- and B-lymphocytes (reviewed in Pfenniger et al., 2013; Glass et al., 2015).
Vascular alterations are invariably involved in diseases of the central nervous system, including stroke, traumatic injuries, and neurodegenerative disorders. The specific parenchymal connexins in neurons and glial cells are introduced in section VIII handling on the role of connexins in cerebrovascular and retinovascular disease.
Transcriptional control of Cx40 and Cx43 genes is known to be mediated by several transcription factors, including Sp1 and Sp3 for Cx40 and Sp1/Sp3 and AP1 for Cx43 (Teunissen and Bierhuizen, 2004). Cx43 transcription is promoted by tetradecanoyl phorbol acetatevia activation of c-jun and c-fos that dimerize to form AP1. Many other signals affect Cx43 expression including thyroid and parathyroid hormones, estrogens (see section V.B.1), prostaglandin E2, and signaling via Ras, Wnt1, and cAMP pathways. In tumor cells, connexins like Cx43 and others are often downregulated by hypermethylation of the gene promoters (reviewed in Vinken, 2016). Histone acetylation has the opposite effect and stimulates the expression of Cx43 and Cx45, as induced by chemical inhibitors of histone deacetylase (HDAC) enzymes such as sodium dibutyrate (Hattori et al., 2007). Other HDAC inhibitors like suberoylanilide hydroxaminic acid upregulate Cx37 but downregulate Cx40 in cardiodystrophic mice (Colussi et al., 2010). The nonspecific connexin channel inhibitor 18-α-glycyrrhetinic acid and carbenoxolone have been reported to act via a decreased transcription (Guo et al., 1999; Herrero-González et al., 2009; Wang et al., 2009) (see section III.A).
A major point of posttranscriptional control is mediated by microRNAs (miRNAs), which may suppress translation or cleave mRNAs and thereby affect mRNA stability. For the cardiovascular connexin family, most evidence is available for Cx43 and Cx40. Cx43 targeting miRNAs bind to the 3′-UTR region in the mRNA (reviewed in Vinken, 2016) and a prominent example is miR-1. When overexpressed in rat heart, miR-1 exacerbates arrhythmogenesis by reducing Cx43 expression as well as expression of Kir2.1 K+ channel subunits (Yang et al., 2007). Interestingly, miR-1 is overexpressed in individuals with coronary artery disease, whereas it is downregulated in patients with tetralogy of Fallot (Wu et al., 2014). Other miRNAs that downregulate Cx43 and lead to arrhythmogenesis are miR-17-92 and miR-130a (Danielson et al., 2013; Osbourne et al., 2014) (see also section V). Some miRNAs promote connexin expression, such as miR-208a that stimulates cardiac Cx40 expression (Callis et al., 2009). Adenylate/uridine-rich elements and RNA-binding proteins also affect mRNA stability, but they have been less characterized in a cardiovascular context (reviewed in Salat-Canela et al., 2015).
B. The Life Cycle from Connexins to Channels and Back
The connexin life cycle encompasses various steps, including 1) connexin trafficking and formation of HCs, 2) GJ assembly, 3) formation of GJ plaques, and 4) disassembly of GJs by internalization and degradation. These various steps are introduced below; their modulation is discussed further in section II.F (for reviews, see Laird, 2006; Salameh, 2006; Thevenin et al., 2013).
1. Connexin Trafficking and Formation of Hemichannels.
After biosynthesis in the ER, connexins are transported by forward trafficking to the plasma membrane via the secretory pathway involving the Golgi apparatus. Most connexins follow this pathway but trafficking for Cx26 (a connexin found in liver, cochlea, Schwann cells, and oligodendrocytes) has been reported to be Golgi dependent (Thomas et al., 2005) as well as independent (Martin et al., 2001). Hemichannels are hexameric connexin assemblies that are formed by oligomerization that starts in the endoplasmic reticulum, continues in the Golgi apparatus, and stabilizes in the trans-Golgi network (Laird, 2006). Some connexins like Cx43, the predominant cardiovascular connexin, and Cx46 (an eye lens connexin) appear monomeric in the ER/Golgi and only hexamerize in the trans-Golgi network (Musil and Goodenough, 1993; Koval et al., 1997). In Cx26, the smallest and best characterized connexin in terms of atomic structure (3.5 Å resolution based on X-ray crystallographic analysis), the subunit interactions that lead to oligomerization involve the outer half of transmembrane helices TM2 (a pore lining domain) and TM4 (a membrane facing domain) as well as the ELs (Maeda et al., 2009). Additionally, amino acids in the CL, close to the transition into TM3 are also involved (Smith et al., 2012). Arg-184 in Cx26 (EL2) is a crucial amino acid that, when mutated, leads to disturbed oligomerization as do mutants of Arg-75 (EL1, just before transitioning into TM2) (Maeda and Tsukihara, 2011); both Arg are well conserved in other connexins.
Vesicles containing HCs, at least those composed of Cx43, are transported along microtubules and actin filaments to the plasma membrane, and tethering of the microtubule plus ends at the adherens junction proteins promotes the plasma membrane delivery process (Shaw et al., 2007). Connexins undergo various posttranslational modifications, including extensive regulation via phosphorylation (see section II.F). Two kinases are involved in connexin trafficking, PKA and PKB (Akt kinase). Increased cyclic AMP and subsequent PKA activation with Cx43 phosphorylation at Ser-364 and Ser-365 promotes trafficking and hence GJ formation (Paulson et al., 2000; Shah et al., 2002); PKB/Akt phosphorylation at Ser-373 promotes trafficking through involvement of 14-3-3 proteins (Park et al., 2007) (a more detailed discussion is given in section II.F).
Once in the plasma membrane, HCs are kept in a closed state by two mechanisms that relate to the membrane potential and the Ca2+ concentration inside and outside the cell. HCs are typically closed at negative membrane potential (Vm) as well as by the 1–2 mM extracellular Ca2+ concentration ([Ca2+]e) (Trexler et al., 1996; Contreras et al., 2003; Verselis et al., 2009). Also in the ER, the millimolar [Ca2+] inside keeps the HCs closed to avoid ER Ca2+ store leakage. Interestingly, GJ formation discussed next is also dependent on [Ca2+]e and involves E-cadherin (Jongen et al., 1991).
2. Gap Junction Channel Assembly.
The next step in the connexin lifetime consists of the head-to-head docking of two HCs from adjacent cells to assemble as a GJ channel. It is currently not known whether and to what extent a separate HC pool exists that is not destined to become incorporated into GJs. Hemichannels on their way to GJs have been proposed to move on lipid rafts until they meet another HC to dock with Schubert et al. (2002). HC docking involves complex interactions between the ELs of the apposed proteins, resulting in the formation of a sealed conduit between the HC heads, located in the 40-Å-wide gap area that separates the two plasma membranes. At this occasion, the two interacting HCs open by the opening of a loop gate (Bukauskas et al., 1995; further discussed under section II.E). Based on 3.5 Å resolution X-ray crystallographic data from Cx26 and work from other authors referred to below, the molecular details of EL interactions can be summarized as follows. First, the ELs within a HC form an anti-parallel β sheet stabilized by three disulfide bridges between Cys residues conserved in all connexins, which connect EL1 and EL2 from each single subunit by intramolecular interactions, thereby maintaining a rigid tertiary structure that favors docking (see Fig. 9 in Foote et al., 1998; reviewed in Sosinsky and Nicholson, 2005; Yeager and Harris, 2007; Maeda and Tsukihara, 2011). Second, based on the Cx26 crystal structure, the interaction between the ELs of opposed HCs is thought to occur such that the EL protrusions from an HC fit into the valleys between ELs of an HC at the opposite side, i.e., each connexin subunit can interact with two subunits of the opposed connexon, forming a β barrel that seals the inside of the channel from the extracellular space (Foote et al., 1998; Yeager, 1998; Yeager and Harris, 2007; Maeda et al., 2009). The inner wall of this junctional structure is formed by the interaction of opposed EL1 domains of each connexin subunit, whereas the outer wall is formed by interacting EL2 domains of each connexin (illustrated in Fig. 1 in Riquelme et al., 2013). Based on a rotation of 30° between the ELs of opposed HCs, an alternative scenario whereby EL1 interacts with an opposed EL2, and EL2 with an opposed EL1, has also been suggested (illustrated in Fig. 4 in Perkins et al., 1998). The EL sequences are highly conserved between different connexins, and data from Cx26 indicate that intercellular EL1-EL1 interactions involve hydrogen bond formation between a highly conserved Asn-54 at one side with Leu-56 at the opposite side, as well as between opposed Gln-57 residues. EL2-EL2 interactions involve hydrogen bonds and salt bridges formed between Lys-168, Asp-179, and Thr-177 from one side and with Asn-176 from the opposite side (Maeda et al., 2009) (Fig. 1).
3. Formation of Gap Junction Plaques.
GJ channels organize in so called plaques that have a size that ranges from just visible by confocal microscopy up to several micrometers in length and contain hundreds to thousands of GJ channels (Klaunig and Shi, 2009) separated by a center-to-center distance of ∼10 nm (Goodenough and Revel, 1970; McNutt and Weinstein, 1970). Very little is known on how exactly GJ channels interact with each other to be arranged in a densely packed well-organized array of a GJ plaque (atomic force microscopy images of gap junction plaques can be found in Müller et al., 2002; Yu et al., 2007). An ill-defined minimum number of channels or channels per plaque is necessary to obtain electrical or dye coupling (Palacios-Prado et al., 2009). Surprisingly, only a limited number of channels in the order of 1% are proposed to be open in a plaque [Cx36 (Marandykina et al., 2013); Cx57 (Palacios-Prado et al., 2009)]. The inclusion of newly assembled GJs in a plaque is an organized event that involves interactions with the zonula occludens 1 (ZO-1) scaffolding protein, which links Cx43 to F-actin. ZO-1 has three PDZ domains and Cx43, as most other connexins, interacts with the PDZ2 (see Thevenin et al., 2013 for more details). Exceptions are Cx36 (a connexin found in neurons and the endocrine pancreas), which interacts with the PDZ1 domain (Li et al., 2004c), Cx45 (a heart and skin connexin) that interacts with all three PDZ domains (Kausalya et al., 2001), and Cx32 (a liver connexin also found in Schwann cells), which interacts with ZO-1 domains other than PDZ or indirectly via intermediate partners (Kojima et al., 2001; Li et al., 2004b). Cx43 interacts with PDZ2 via a DLEI sequence located at the C-terminal end (see Fig. 1), with the last four amino acids being crucial; removal of the last amino acid (Ile-382) prevents the interaction (Jin et al., 2004). Cx43 HCs are linked to ZO-1 proteins at the periphery of a GJ plaque in a zone called the perinexus that is distinct from the centrally located “nexus” structure that contains the GJs (Rhett et al., 2011; Rhett and Gourdie, 2012). The dynamics of plaque formation is incompletely understood; it involves accretion of newly formed GJs at the outer edge of the plaque, whereas old GJs are removed from the center (Gaietta et al., 2002; Lauf et al., 2002), probably by endocytosis of small, or even large, circular double membrane vesicles called “annular junctions” or “connexosomes” (Archard and Denys, 1979; Jordan et al., 2001; Falk et al., 2009). Outer rim accretion is a two-step process whereby HCs assemble as GJs and GJs incorporate into plaques; the connexin−ZO-1 interaction is crucial in this process. When Cx43 is tagged with GFP at its CT end, a domain involved in interaction with ZO-1, GJ plaques grow significantly larger (Hunter et al., 2003). Additionally, adding a peptide composed of the last nine amino acids of the Cx43 CT (Arg-374–Ile-382 further called CT9, see Fig. 1) and linked to an antennapedia internalization sequence to make it membrane permeable (called αCT1) has the same plaque increasing effect (Hunter et al., 2005). This and other work from the Gourdie group, making use of various approaches, including duolink (proximity ligation) protein-protein interaction studies, demonstrates that HCs are linked to ZO-1 and that disruption of this linkage is crucial in the process of HC incorporation and GJ assembly into the plaque (reviewed in Palatinus et al., 2012; Rhett et al., 2013). Given the multitude of scaffolding and cytoskeletal links to the CT, other proteins and structures such as microtubules (Toyofuku et al., 1998; Giepmans et al., 2001) and myosin VI (Waxse et al., 2017) may also be involved. At least two kinases are involved in GJ assembly and incorporation in GJ plaques: PKB/Akt and casein kinase (CK)1. PKB/Akt phosphorylation of Ser-372 or -373 helps in loosening the binding of Cx43 to ZO-1 (Chen et al., 2008); CK1 phosphorylation of Ser-325, Ser-328, and Ser-330 promotes GJ/plaque assembly (Cooper and Lampe, 2002) and promotes dye coupling and GJ channel opening (Lampe et al., 2006). Of note, ZO-1 interactions with the Cx43 CT are not necessary for HC/GJ formation (Fishman et al., 1991; Dunham et al., 1992). ZO-1–CT interactions are modulated by various influences: 1) c-Src interacts with the Src homology domain 3 (SH3) domain on the Cx43 CT [Pro-274–Pro-284 (Kanemitsu et al., 1997), see Fig. 1], and this displaces the ZO-1–CT interaction that is located ∼100 amino acids away from the SH3 domain (Sorgen et al., 2004); 2) ZO-1–CT interactions decrease upon acidification, and this favors c-Src binding to CT (Duffy et al., 2004); and 3) ZO-1–CT interactions increase upon isolating cardiomyocytes from their native environment and are thus involved in remodeling (Barker et al., 2002).
4. Gap Junction Disassembly: Internalization and Degradation.
Several possible scenarios exist for GJ removal. Early work suggested dispersal of junctional particles in the plasma membrane, away from the junction (Lane and Swales, 1980). It has also been proposed that GJs may split into reseparated HCs according to an unzippering scenario. GJ unzippering has been suggested as a mechanism for peptide inhibitors of GJ channels like Gap26 or Gap27 (see section III) (Berthoud et al., 2000), but there is currently no hard evidence to support such a mechanism. Possibly, HCs are also internalized: making use of a fluorescently labeled peptide identical to a Cx43 EL1 sequence and CFP-tagged Cx43, Dermietzel et al. (2003) demonstrated uptake of double-labeled structures, indicating inward trafficking of HCs (Dermietzel et al., 2003). However, the currently best characterized uptake scenario consists of the uptake of an entire GJ plaque (or parts of it), which then appears in one of the cells as an annular junction (Jordan et al., 2001; Piehl et al., 2007; Falk et al., 2009). The uptake process can occur by clathrin (Gumpert et al., 2008)- or caveolin-dependent (Schubert et al., 2002; Lin et al., 2003b) endocytosis (reviewed in Thevenin et al., 2013). Clathrin-dependent uptake involves several clathrin adapter proteins such as adaptor protein complex-2 that interact with Tyr-based sorting signals located on the CT tail (in particular Y265AYF, Y286KLV, and others for Cx43; see Fig. 1) and disabled2. Dynamin2, a GTPase, facilitates the inward pinching-off of plasma membrane vesicles, whereas myosin-VI provides vectorial inwardly directed transport. Further breakdown can occur via lysosomal or proteasomal pathways (Laing and Beyer, 1995; Laing et al., 1997,1998; Qin et al., 2003, reviewed in Salameh, 2006). Ubiquitination is central for proteasomal breakdown, but proteasomal breakdown of Cx43 can occur in the absence of ubiquitination (Su et al., 2010; Su and Lau, 2012); ubiquitination can furthermore be involved in endosomal/lysosomal breakdown. Ubiquitination occurs at Lys residues, of which Cx43 has 23; 20 of them are located on the CL and CT, which contain the more likely ubiquitination sites. The large GJ structures are not typical targets for proteasomal breakdown, but Girão et al. (2009) reported that a Cx43-ubiquitin fusion protein present in HCs and GJ plaques was internalized by a process that involves the ubiquitin ligase Nedd4 and the endocytic adapter protein Eps15. Nedd4 interaction with Cx43 occurred at a Pro-rich motif in the CT that overlaps with the Tyr-286 based sorting signal (Y286KLV) (Fig. 1) (Leykauf et al., 2006; Girão et al., 2009). Mutation of Tyr-286 and of downstream Val-289 increases the Cx43 half-life and stability of the GJs (Thomas et al., 2003; Catarino et al., 2011). Ubiquitination may also play a role in the endosomal/lysosomal pathway where it may target endosomes for lysosomal breakdown via the Tsg101 sorting protein that also binds to Cx43 (Auth et al., 2009; Leithe et al., 2009). In addition to proteasomal and lysosomal degradation, growing evidence indicates involvement of autophagic degradation of connexins as well (Lichtenstein et al., 2011; Bejarano et al., 2012; Falk et al., 2012; Fong et al., 2012; Iyyathurai et al., 2016). These various removal/internalization pathways offer interesting, yet unexplored, handles to modulate GJ size (Beyer and Berthoud, 2002).
C. Non-Channel Functions of Connexins
Connexins are best known for their function as channels, GJs and HCs. However, connexin proteins are also endowed with channel-independent functions. Many, but not all, of the below described non-channel effects link to the CT tail of Cx43, which is a hub for interactions with other scaffolding or signaling proteins (Giepmans, 2004; Hervé et al., 2012) and also the site by preference for modulatory phosphorylation events. The list of non-channel functions includes connexin interactions with cytoplasmic signaling molecules, connexin-linked sequestering of transcription factors, CT-migration to the nucleus, and adhesion aspects linked to the mechanical cell-cell connection provided by GJs.
In terms of interaction with cytoplasmic signaling molecules, there is evidence that connexins interact with apoptotic factors thereby influencing cell death. Cx26 and Cx43 colocalize with the Bcl-2 proteins Bak, Bcl-xL, and Bax in the cytoplasm of human breast and colorectal cancer cells (Kanczuga-Koda et al., 2005a,b). Recent evidence demonstrates a more direct interaction of Cx43 with apoptosis signal-regulating kinase 1 (Giardina et al., 2007). Another example comes from the glioma brain tumor field. In these tumors, the levels of Cx43 decrease with increasing malignancy (Sin et al., 2012). The inverse relation between tumor cell proliferation and Cx43 expression in part links to the fact that Cx43, in particular the CT tail, exerts a brake on cell proliferation (Zhang et al., 2003). Work from the Tabernero group recently demonstrated that Cx43 inhibits proliferation by interacting with the proto-oncogene Tyr protein kinase c-Src via a CT-located SH3 domain (see Fig. 1), which involves corecruitment of the phosphatase and tensin homolog and CT Src kinase (González-Sánchez et al., 2016; reviewed in (Tabernero et al., 2016). Interestingly, a peptide composed of Ala-266 to Pro-283 [PEP-2 (Gangoso et al., 2014); see Fig. 1] was able to mimic the effect of the full Cx43 protein in limiting the proliferation glioma stem cells (González-Sánchez et al., 2016); more recent work demonstrated that PEP-2 was able to reduce migration, invasion, and survival of primary glioma stem cells isolated from human glioma tumor samples (Jaraiz-Rodriguez et al., 2017).
In terms of signaling to the nucleus, there are two possibilities: either the CT is involved in sequestering cytoplasmic transcription factors in a direct/indirect manner, or the CT itself translocates to the nucleus. The transcription factor ZONAB (ZO-1-associated nucleic acid binding protein) is an example of the first possibility. ZONAB associates, through its ZO-1 binding partner, with the glial connexins Cx30, Cx43, and Cx47 (Penes et al., 2005; Li et al., 2008b). ZONAB not linked to ZO-1 migrates to the nucleus where it acts as a transcription factor that binds to promoter sequences containing an inverted CCAAT box (Balda and Matter, 2000) and modulates the cell cycle, mostly resulting in increased proliferation (reviewed in Zahraoui, 2004). The ZONAB link to connexins and cell proliferation is thus indirect and based on evidence from the tight junction field; in any case ZONAB-sequestering by Cx30, Cx43 and Cx47 may be involved in inhibition of cell proliferation by these connexins. Next to ZO-1 sequestering of ZONAB, Cx43 also interacts with ZO-2 (Singh et al., 2005), which sequesters the transcription factors c-jun, c-fos and CCAAT enhancer binding protein (Betanzos et al., 2004), again resulting in decreased cell cycle activity and proliferation (Tapia et al., 2009). Another example of connexin-sequestering comes from the cell-cell adhesion molecule β-catenin. Cx43 interacts with β-catenin (Ai et al., 2000) and in this way influences Wnt/β-catenin signaling, which acts to promote cell cycling and proliferation. Thus, Cx43─β-catenin interaction reduces β-catenin availability, inhibiting its migration to the nucleus and its subsequent interaction with transcription factors, thereby decreasing cell proliferation (MacDonald et al., 2009). Direct migration of the Cx43 CT to the nucleus has also been implicated in the Cx43-suppressive effect on cell proliferation despite the fact that the CT does not contain a known nuclear target sequence (Dang et al., 2003).
Connexins are also endowed with cell adhesion functions (Lin et al., 2002; reviewed in Prochnow and Dermietzel, 2008). The Kriegstein group demonstrated that the radial migration of excitatory neuronal precursor cells along radial glial cells during cortical development depended on adhesive properties provided by GJs composed of Cx26 and Cx43 (Elias et al., 2007). Follow up work showed this was also the case for the more complex migration pathway of inhibitory neuronal precursors, which come from lateral and make a switch in radial direction upon interacting with radial glia (Elias et al., 2010; reviewed in Elias and Kriegstein, 2008). Work from the Naus group demonstrated that adhesion-linked connexin properties also play a role in the migration and invasion of glioma cells (Sin et al., 2016; reviewed in Matsuuchi and Naus, 2013; Naus et al., 2016). Cx43-mediated adhesion effects have also been reported for the spreading of B-lymphocyte adhesion when cultured on a substrate (Machtaler et al., 2011). Forced expression of Cx43 in prostate cancer cells was furthermore found to promote metastasis to the bone, which strongly expresses Cx43 (Lamiche et al., 2012). This suggests that the docking of two apposed HCs and/or the function of GJs may play a role in tumor metastasis to specific organs. Breast tumor carcinoma cells express Cx43 and frequently metastasize to the brain; recent work has provided evidence that breast metastatic carcinoma cells take advantage of GJ communication with Cx43-expressing astrocytes, demonstrating that the role of connexins in metastasis definitely involves channel functions (Chen et al., 2016).
Last but not least, alterations in all aspects of the connexin life cycle (see section II.B) may in some way be linked to non-channel effects. For example, changes in connexin expression or the presence of mutant connexins may affect the trafficking of other proteins that follow the secretory pathway and thereby influence intracellular vesicle trafficking and the release of exosomes, which also contain Cx43 (Soares et al., 2015). A striking example comes from the heart, where the loss of Cx43 expression leads to reduced Na+ currents in ventricular (Danik et al., 2008; Jansen et al., 2012) and atrial cardiomyocytes (Desplantez et al., 2012a). Similar findings of a reduced Na+ current, further complicated by ventricular fibrillation, were observed in a cardiac knock-in model of Cx43 that lacked the last five amino acids in the CT [Cx43D378stop mutation (Lübkemeier et al., 2013)]. Work of the Delmar group in this model demonstrated that, although GJ coupling is normal, NaV1.5 delivery to the plasma membrane was impaired in a microtubule-dependent manner (Agullo-Pascual et al., 2014b). Omission of the five last Cx43 CT amino acids was found to limit the capture of the microtubule plus end tracking protein end-binding-1 at the ID, resulting in impaired cargo delivery to this location (reviewed in Leo-Macias et al., 2016).
Further reviews on non-channel functions of connexins can be found in Giepmans (2004), Jiang and Gu (2005), Dbouk et al. (2009); Hervé et al. (2012), Vinken et al. (2012), Zhou and Jiang (2014); Table 1 of Zhou and Jiang (2014) gives an overview of channel-independent effects.
D. Permeability
The structure of HCs and GJ channels composed of Cx26 has been studied in much detail, making use of three-dimensional X-ray crystallographic analysis at 3.5 Å resolution (Maeda et al., 2009; reviewed in Maeda and Tsukihara, 2011). These studies suggested that the smallest pore diameter of the Cx26 channel is ∼14 Å. As a consequence, Na+, K+, and Cl− can pass through in the presence of their hydration shell [diameters 3.58, 3.31, and 3.32 Å, respectively (Volkov et al., 1997)]. The low-pass cut-off for a 14-Å-wide pore is in the order of ∼1.2 kDa MM based on an empirically derived relation between diameter and molecular weight for globular proteins (Erickson, 2009). Based on size exclusion studies with Alexa dyes, the pore diameter of Cx43 channels was estimated to be in the order of 14.8 Å (Weber et al., 2004), giving a calculated ∼1.4 kDa MM cut-off. For Cx37, size cut-off based on polyethylene glycols was in the 6.8–8 Å range (Gong and Nicholson, 2001). In addition to size, permeation also depends on charge. Although the narrowest diameter of the pore is much wider than hydrated Na+ or Cl−, the permeability for these ions can be different, indicating charge selectivity, even for atomic ions (Veenstra et al., 1995). For larger charged molecules, which are more likely to interact with the channel wall, charge selectivity is even more pronounced and sometimes very different between different connexins. For example, negatively charged molecules like carboxyfluorescein (8.2 Å, charge −2) or Lucifer yellow (9.9 Å, charge −2) permeate better through GJ channels composed of Cx43 than those composed of Cx40. By contrast, positively charged molecules like ethidium bromide (10.3 Å, +1) permeate better through Cx26-based channels (Kanaporis et al., 2011). Despite the importance of size and charge, there are no obvious rules to explain the permeation of metabolic or signaling molecules, suggesting that still unknown interactions (electrostatic, van der Waals, or covalent in nature) occur within the channel pore (Weber et al., 2004; reviewed in Harris, 2007). As a result, channel permeation does not occur along a free diffusion scenario but based on selectivity properties that are still poorly understood (Ek-Vitorin and Burt, 2013). For example, Cx26 channels are more or less equally permeable to cAMP (MM 329) and IP3 (MM 420) despite the large difference in charge (neutral and −6, respectively) (Hernandez et al., 2007). ATP (MM 507, charge −4) permeates a 300-fold better through Cx43 channels than through Cx32 channels (Goldberg et al., 2002), a huge difference probably difficult to explain on pore size only. For cAMP, the permeation sequence is Cx43 > Cx26 > Cx45 = Cx32 (Wang and Veenstra, 1997; Bedner et al., 2003, 2006). Additionally, the permeability for fluorescent dye molecules does not always go hand in hand with the channel electrical conductance properties (fully open state versus substrates; see section II.E) (Brink et al., 2006; Eckert, 2006) and high conductance channels like those formed by Cx37 display poor permeation of fluorescent dyes (Veenstra et al., 1994, 1995). Clearly, connexin channels display connexin-specific permeability profiles, and more studies are needed to have a better understanding of both the exact pore size (based on structural and functional approaches) and the nature of the interactions of the permeant with the channel. This is especially relevant for the larger permeating substances that are more likely to interact with the channel pore-lining TM1 and TM2 domains compared with atomic ions that have more space available and therefore are able to diffuse more freely with less interactions with the channel wall-lining residues; when the diameter of the permeant comes in the range of the pore diameter, interactions with the channel wall will rather determine the permeability profile.
Connexin channels can also be formed by different connexins, resulting in heteromeric or heterotypic channels. HCs can be homomeric or heteromeric (composed of different connexins). Two homomeric HCs make a homotypic GJ channel. A heterotypic GJ channel can be formed by two HCs each composed of a different connexin or by two heteromeric HCs. Two heteromeric HCs may also form a homotypic GJ channel when each subunit pairs with a subunit composed of the same connexin isoform (reviewed in Koval et al., 2014; the various configurations are illustrated in Fig. 1 of Mese et al., 2007). The most studied heteromeric/heterotypic GJs are formed by the cardiovascular connexins Cx37, Cx40, Cx43, and Cx45, which can form heteromeric (He et al., 1999; Beyer et al., 2013) and heterotypic GJs (Lin et al., 2014). The compatibility of different connexins to form heteromeric/heterotypic channels is determined by motifs located at the CL-TM3 transition and left half of EL2. Importantly, heteromericity largely influences GJ permeability for metabolites or signaling molecules, thereby allowing the fine-tuning of the repertoire of substances exchanged between cells (Harris, 2007). Interestingly, heterotypic channels display electrical rectification behavior that may facilitate transport and electrical conduction in a preferential direction (Rackauskas et al., 2007; reviewed in Harris, 2002).
Connexin channels are often called aqueous pores, meaning they allow ions in aqueous solution to pass through the pore surrounded by their hydration shell, as mentioned earlier. In principle, water is able to pass through the channel but only little information is currently available demonstrating direct water flow through the channels. In the lens, GJs composed of Cx46 and Cx50 have been reported to be involved in water flow (Gao et al., 2011). Work on HCs seemed to confirm this, based on cell swelling observed in cell expression systems [Cx43 (Quist et al., 2000); Cx46 (Paul et al., 1991)]. However, more detailed analysis performed on HCs composed of Cx30 or Cx43 has demonstrated they are not water permeable (Hansen et al., 2014). Taken together, connexin channels may be involved in water fluxes, but this is probably the result of ionic flow through GJs or HCs with subsequent osmotic water flow through other pathways, e.g., via aquaporins.
E. Gating
Connexin channels are modulated by numerous influences that link to the cell state or internal/external influences, which most commonly link to voltage, pH, Ca2+ concentration, phosphorylation state, and redox state. Several conditions, especially inflammatory conditions, strongly affect connexin channels. Connexin channel permeability expresses the degree of permeation of noncharged permeants, whereas conductance expresses the degree of permeation of charged permeants, i.e., the ease of current passage. Macroscopic conductance is the conductance of an array of channels; modulatory influences alter the macroscopic conductance, which can be caused by changes in the number of channels present, the conductance of single channels, or their gating, i.e., opening and closing activities. Single channel conductance (γ) and gating properties are determined in voltage-clamp experiments, dual cell voltage-clamp for GJ channels and single cell patch-clamp experiments for HCs. These experiments further bring up important information on other biophysical properties like the rise time for opening/closing transitions and the reversal potential, i.e., the potential at which the current reverses direction. Because connexin channels have no selectivity for atomic ions like Na+, K+, Ca2+, and Cl−, the reversal potential is in most cases ∼0 mV. The single channel conductance is a fundamental property that differs among the different members of the connexin protein family; hence it can be used to determine what connexin is recorded from (an overview of single channel conductances of HCs and GJs is given in Table 3 of Sáez et al., 2005). Below follows a brief overview on the gating of GJs and HCs by voltage, Ca2+, and pH. Connexin channel gating is a very complex field, in part because the gating, as well as other biophysical properties, differ substantially between different connexins. For that reason, observations in channels composed of Cx26, Cx32, Cx45, or Cx46 to name some, are not necessarily true for Cx43, which we keep as the prototypic example case for this review. Excellent entries into the connexin gating subject can be found in Bukauskas and Verselis (2004), Sáez et al. (2005), González et al. (2007), Moreno and Lau (2007), Ek-Vitorin and Burt (2013), Fasciani et al. (2013), Oshima (2014), and Oh and Bargiello (2015). What follows below is an overview of the most salient aspects of gating, highlighting some proposed structural correlates.
1. Slow Loop Gating and Fast Gating.
HCs are normally closed but open upon docking with an apposed HC by a process of loop gating (Bukauskas et al., 1995; Trexler et al., 1996; Contreras et al., 2003), which involves movements of residues at the border of TM1/EL1 [in particular at position 43–50 for Cx50 (Verselis et al., 2009)] and is characterized by slow gating (transition time ≥10 ms) (reviewed in Bukauskas and Verselis, 2004). Loop gating involves large conformational changes, in particular in the first half of EL1 (Bargiello et al., 2012). Loop gating is associated with various aspects, including 1) the docking process of two HCs per se, 2) the change in voltage sensed by the two HCs upon assembling into a GJ, and 3) the change in [Ca2+] at the external side of the HC upon GJ assembly. The first aspect links to the interaction of the ELs during docking, which can be seen as ligand-induced gating (Trexler et al., 1996). The second aspect links to the fact that the negative Vm sensed by the HCs, which keeps them closed, disappears upon formation of a GJ channel between two cells with equal (or approximately equal) Vm. The third aspect links to the lowering of [Ca2+] sensed by the outer half of the HC upon docking and sealing-off the channel interior from the extracellular space where [Ca2+] is in the millimolar range (see more in section II.E.4). GJs start to close (with slow kinetics taking hundreds of milliseconds to seconds) when the junctional potential difference (Vj), i.e., the difference in Vm between the cells connected by the GJ, starts deviating from 0 mV (reviewed in Bukauskas and Verselis, 2004; Palacios-Prado and Bukauskas, 2012). As a result, the relation between Vj (abscissa) and GJ macroscopic current (ordinate) is a convex-up bell-shaped curve centered around the ordinate, with channel closing at strongly negative or positive voltages (see Fig. 2 in Bukauskas and Verselis, 2004). Single channel analysis has demonstrated that Vj closure of GJ channels is, in contrast to slow loop gating, mediated by fast gating (transition time <2 ms) by a gate that is different from the loop gate (Bukauskas and Verselis, 2004). Slow gating is also involved in the closing process, in particular at the top of the bell-shaped GJ voltage dependence curve (around Vj = 0 mV, where gating starts closing fully open GJ channels) as well as at edges where full closure of the channels is at stake (see Fig. 2 in Bukauskas and Verselis, 2004). As alluded to above, HCs not incorporated into GJs can be opened by depolarizing or positive membrane potentials (Sáez et al., 2005); HC gating also displays slow and fast transitions just like GJs (Contreras et al., 2003). The majority of the gating state transitions of GJs/HCs is slow and only transitions to a residual (substrate) conductance, characterized by a γ that is lower than the fully open channel (∼60 pS for Cx43 HCs), are fast. The slow gate always opens with depolarizing or positive voltages (Vm); the fast gate opens with positive Vm for Cx32 but closes with positive Vm for Cx26. As a result, Cx32 HCs increase their open probability with increasing positive voltages; in contrast, Cx26 HCs open with positive voltages but start to close again at increasing positivity, displaying a typical bipolar voltage-dependence (González et al., 2007; Fasciani et al., 2013). Cx43 HCs reportedly also display a bipolar voltage-dependence; however, although slow gate opening clearly increases with above threshold positive membrane potentials, the behavior of the fast gate is more difficult to study because of the paucity of closing events to the residual state.
At the electrical level, GJs behave, at least in a first approximation, as a linear assembly of two apposed HCs; their conduction is half the conduction of the corresponding HC and the gates appear to retain their fast and slow gating polarities in GJs as in HCs (González et al., 2007). A factor that complicates things is the fact that the electrical field sensed by each of the two HCs in a GJ depends on the closed/open state of these HCs (Paulauskas et al., 2012). Of note, GJs do not always behave as a superposition of two HCs, and the responses of both channel types to chemical signals can sometimes be opposite, as discussed below in the context of [Ca2+]i influences on channel gating. The subconductance state is considered the ground state for electrical gating mediated by fast gating (Valiunas et al., 1997; Ek-Vitorin and Burt, 2013). The ground state for “chemical gating” is the fully closed state and transitions to the fully open state are characterized by slow gating (Ek-Vitorin and Burt, 2013). Two of the most studied examples of chemical gating concern the influence of pH and Ca2+ on GJs/HCs, discussed below.
2. Intracellular pH Effects on Gating: The Role of C-Terminal Tail-Cytoplasmic Loop Interaction.
GJs close with intracellular acidification (pH 5–6), as occurs in the context of ischemia. Interestingly, when the CT of Cx43 is truncated at Ala-257 (last 125 amino acids, including Ala-257), GJ closure upon acidification disappears (Liu et al., 1993). Moreover, re-expression of the truncated CT part rescues pHi sensitivity, indicating involvement of the truncated CT part in channel closure (Morley et al., 1996). Based on these observations, a model of Cx43 CT-CL interaction was proposed whereby the CT binds to a receptor structure, the L2 domain located in the second (CT directed) half of the CL (Asp-119 to Lys-144; Fig. 1). This intramolecular interaction was proposed to close the channel according to a ball-and-chain scenario, analogous to the ball-and-chain mechanism for the closure of Na+ and K+ channels. The binding of the CT to the L2 domain is facilitated under low pH conditions, as a result of a higher α-helical order in the L2 domain (Duffy et al., 2002); thus, CT-CL interaction and its consequences are strengthened by acidosis. Interestingly, CT truncation at residue 257 removed the residual state of the channel, suggesting that the ball-and-chain CT-CL interaction may act as a fast gate (Moreno et al., 2002). Even more interesting were the observations that addition of the L2 domain as an exogenous peptide decreased the transitions from the fully open state to the residual state and increased the open time of the GJ channels (Seki et al., 2004). Mutation of His-142, located in the L2 domain, to Glu-142 has a similar effect (Shibayama et al., 2006a). Thus, intramolecular CT-CL interactions close GJs upon acidification and addition of the L2 peptide prevents this closure. Substantial data are available in terms of the location of the CT domains that interact with the L2 domain. Morley et al. (1997) reported (based on deletion studies) that the regions Cys-260–Asn-300 and Arg-374–Ile-382 from the CT domain were crucial (Morley et al., 1997) (Arg-374–Ile-382 corresponds to CT9, see Fig. 1). Follow up work demonstrated that a peptide composed of Cys-271–Lys-287 (called C271-K287 peptide in Fig. 1) could prevent acidification-induced uncoupling (Calero et al., 1998). Duffy et al. (2002) showed that three CT peptides with sequences Cys-271–Lys-287, Asp-336–Gly-350, and Lys-346–Asp-360 (mouse sequences) all interacted with L2 in surface plasmon resonance experiments (Duffy et al., 2002). The first two of these peptides, respectively, cover the SH3 domain (Pro-274–Pro-284) and an α-helical domain (Asp-340–Ala-348); the third one is situated between the α-helical domain and the CT9 sequence (see Fig. 1 for the location of these domains). Of note, the Cx43 CT contains two α-helical domains, Ala-315–Thr-326 and Asp-340–Ala-348 (illustrated as two yellow colored domains in Fig. 1), with the rest of the CT being an intrinsically disordered structure (Sorgen et al., 2004). Combined surface plasmon resonance and NMR evidence has indicated that CL interaction with the CT involves the second α-helical domain and a region of the last 19 amino acids of the CT that includes the CT9 domain (Hirst-Jensen et al., 2007). In line with this, RXP-E peptide, discovered through a phage display search for high affinity Cx43 CT binders, was reported to interact with residues 343–346 (within the second α-helical CT domain) and 376–379 (within the CT9 domain; see Fig. 1); this peptide partially prevented acidification- and octanol-induced GJ closure (Shibayama et al., 2006b).
Evidence for involvement of CT-CL interaction in connexin channel gating is also available for Cx46, Cx40, and Cx26. Recent evidence indicates that the G143R mutation in the second CT-directed half of the Cx46 CL acts to strengthen CT-CL interaction (Ren et al., 2013); this results in closure of GJs and opening of HCs, in line with the findings of CT-CL effects on GJs and HCs composed of Cx43 (for details on CT-CL effects on HCs see section II.E.5). For Cx40, CT truncation at residue 248 resulted in the disappearance of the residual conductance (Anumonwo et al., 2001), an effect that is similar to what is found in CT-257 truncated Cx43. Additionally, the Cx43-CT was able to substitute for the Cx40-CT, indicating that heterodomain interactions between Cx40 and Cx43 (two prominent vascular connexins) are also possible (see also Bouvier et al., 2009). Experiments with CT-257 truncated Cx43 further suggest that CT-CL interaction is also involved in GJ inhibition by insulin and insulin-like growth factor (Homma et al., 1998). Additional evidence is available for GJ inhibition by v-Src, where CT truncation at residue 245 was most potent in removing GJ inhibition (Zhou et al., 1999). CT-CL interaction is also involved in HC regulation by [Ca2+]i (see section II.E.5).
For Cx26, which has a very short CT (10 vs. 150 amino acids for Cx43), things are different. Locke et al. (2011) reported CT-Cl interactions at normal pH; upon acidification, protonated aminosulfonates like taurine interact with the CL (second CT-directed half) and thereby disrupt the CT-CL interaction (Bennett, 2011; Locke et al., 2011). Thus, opposite to Cx46, Cx43, and Cx40, for Cx26 it is rather a disruption of CT-CL interaction that acts to close the channels, both GJs and HCs.
Some final remarks to conclude this part on CT-CL interaction and pH effects. As mentioned earlier, CT-CL interactions link to fast gating transitions to the subconductance state, which stands in contrast to the fact that chemical gating is proposed to be mediated by slow gating. It needs to be added that issues concerning the identity and functions of slow and fast gates are not unequivocally set. Additionally, GJ closure with acidification is not the prerogative of CT-CL interactions; direct effects mediated by His residues have also been reported [e.g., His-95 in TM2 of Cx43 (Ek et al., 1994)]. Another possibility for GJ closure relates to the fact that protons can displace Ca2+ from common binding sites, thereby indirectly closing the GJs by an increase of [Ca2+]i (see Peracchia, 2004 and section II.E.5). Of note, a low pHi inhibits both GJs and HCs (Spray et al., 1981; Trexler et al., 1999) and there is also evidence that high pHe promotes HC opening (Schalper et al., 2010). As always, exceptions exist and acidosis due to moderate acid load was recently reported to increase rather than suppress GJ coupling (Swietach et al., 2007).
3. Voltage-Dependent N-Terminal Tail-Linked Gating.
Just as the CT plays an important role in connexin channel gating, the NT domain also plays a forefront role. It has been known from 20 years of work from the Bargiello group that the NT, which is very short (∼20 amino acids for Cx26, ∼22 for Cx32, and ∼13 for Cx43), is involved in Vj voltage sensing (Verselis et al., 1994; Oh et al., 2000; Purnick et al., 2000). Especially position 2 as well as amino acids at the TM1/EL1 border appear to be crucial for voltage sensing in both Cx26 and Cx32 (for Cx26 crucial amino acids are Asp-2 and Lys-41 at TM1/EL1). Work from these authors demonstrated that NT-linked gating involved switching between open and a subconductance state and suggested that the NT moves toward the cytoplasm upon HC closure (Verselis et al., 1994). Moreover, it was concluded that NT movement of a single subunit was sufficient to switch to the subconductance state (Oh et al., 2000), which is in contrast to the cooperative tilting model whereby all subunits move in a concerted manner to open or close the channel (Unwin and Ennis, 1984). Later work by Maeda et al. (2009) based on 3.5 Å crystal structure of Cx26 demonstrated that the NT makes a bend into the cytoplasmic side of the HC and is kept there in a stable position by an interaction that involves Met-34 in TM1 and others in TM1, first half of EL1, and TM2 (Maeda et al., 2009; Harris and Contreras, 2014). This particular “attached” configuration of the NT is hypothesized to determine the 14 Å pore diameter of the open channel. When the voltage-sensing amino acids (especially Asp-2 and Lys-41 for Cx26) are exposed to a large junctional voltage difference, the NT-attachment in the pore wall is destabilized, causing NT movement toward the cytoplasm and narrowing down of the pore to a subconductance state.
Both CT-CL and NT-linked gating has been proposed to be involved in gating but from different contexts that link to chemical (pH) and voltage gating, respectively. It is currently unestablished whether NT- and CT-linked gating have something in common; in that regard, the CL has been proposed as a candidate interaction target for both CT and NT (see e.g., Fig. 14 in Harris, 2001). Another question is how the voltage-sensing amino acids in NT/TM1/EL1 CT could influence CT-CL interaction. Of note, the voltage-sensing domain of the slow gate is currently unidentified. Clearly, connexin channels do not have a defined and unique voltage-sensor domain like the S4 domain of Na+ channels, where four or more positively charged amino acids (Arg or Lys) are each separated by two non-charged amino acids (see Fig. 2 in Moreau et al., 2014). Likewise, the gates of connexin channels are not separate units clearly distinguishable from the voltage sensor (Harris and Contreras, 2014), and it looks like connexins combine their voltage sensing, gating, and perhaps also permeation conduit functions in multipurpose domains.
4. Ca2+-Based Gating–Effects of Extracellular Ca2+.
Plasma membrane HCs not incorporated into GJs are kept in a closed state by several mechanisms that include absence of loop interactions with opposed HCs, the negative Vm, and the 1–2 mM extracellular Ca2+ concentration ([Ca2+]e) (reviewed in Fasciani et al., 2013; Harris and Contreras, 2014). When [Ca2+]e falls, HCs will open as demonstrated in cardiomyocytes (Kondo et al., 2000), blood vessel endothelial cells (De Bock et al., 2012), blood cells (Eltzschig et al., 2006), and glial cells in brain slices (Torres et al., 2012). Almost all connexins display HC opening at low [Ca2+]e, including Cx26 and Cx30 (Lopez et al., 2013, 2016), Cx32 (Gómez-Hernández et al., 2003), Cx37 (Puljung et al., 2004), Cx40 (Allen et al., 2011), Cx43 (Bruzzone et al., 2001; Contreras et al., 2003; Ye et al., 2003; Thimm et al., 2005), Cx46 (Paul et al., 1991; Ebihara et al., 2003), and Cx50 (Zampighi et al., 1999; Beahm and Hall, 2002). In general, HCs start to open when [Ca2+]e decreases below 0.5 mM, but there are substantial differences between different connexins. In addition to Ca2+, Mg2+ also plays a role but Ca2+ exerts a dominant effect (Ebihara et al., 2003). Work from the Lal group with atomic force microscopy elegantly demonstrated that lowering [Ca2+]e increases the Cx43 HC outer pore size (Thimm et al., 2005). In Cx32 (a liver, oligodendrocyte, and Schwann cell connexin), the high [Ca2+]e closure has been traced down to Asp-169 and Asp-178 in the first half of EL2 that together form a ring of 12 Asp residues in the external vestibule of the HC (Gómez-Hernández et al., 2003); this structure is conserved in Cx30 (an astrocyte and skin connexin), Cx46, and Cx43. For Cx46, the gate that closes the channel with millimolar [Ca2+]e has been reported to be situated in extracellular direction relative to the position of Leu-35 in TM1 (Pfahnl and Dahl, 1999). Recent work from the Contreras group, based on computational molecular dynamics simulations of Cx26, has demonstrated that [Ca2+]e closes the HCs by disrupting intersubunit salt bridges between Asp-50 and Lys-61 located in EL1 (Lopez et al., 2016). The authors proposed that Ca2+ binds to Asp-50 and Glu-47 in the same region, which would form a ring of 12 negatively charged residues as suggested for Cx32 (Gómez-Hernández et al., 2003). The Lopez et al. (2016) study also suggested that the actual gate closing the channel is presumably located deeper into the channel. The Barrio group brought up evidence that, for Cx32, HC opening triggered by lowering [Ca2+]e involves transitions to a subconductance state (∼18 pS) and from there to a main conductance state (∼90 pS), with an inverse sequence occurring upon restoration of normal [Ca2+]e (reviewed in Fasciani et al., 2013). Of note, [Ca2+]e may also affect voltage gating as demonstrated for Cx46 (Ebihara and Steiner, 1993; Ebihara et al., 2003; Verselis and Srinivas, 2008). In particular, lowering [Ca2+]e results in the unshielding of negative charges associated with the glycocalyx outside the cell, which attracts positive charge at the inside of the membrane capacitor and acts as a depolarizing stimulus on the voltage sensor. As a result, the voltage threshold for activation is lowered, resulting in a left shift of the voltage activation curve. Moreover, De Vuyst et al. (2006), reported that hemichannel-mediated ATP release triggered by exposure of Cx32 expressing cells to divalent-free solutions (Ca2±- and Mg2+-free) was dependent on [Ca2+]i changes, indicating that lowering [Ca2+]e also influences HC opening by provoking [Ca2+]i dynamics (De Vuyst et al., 2006).
5. Ca2+-Based Gating–Effects of Intracellular Ca2+.
In contrast to [Ca2+]e that only influences HCs, alterations of [Ca2+]i affect both GJs and HCs. We first discuss effects on GJs and subsequently discuss more recently found effects of [Ca2+]i on HCs.
a. Intracellular Ca2+ concentration effects on gap junctions.
GJs close in response to large Vj differences and acidification, but also in response to [Ca2+]i elevation. Rose and Loewenstein (1975) demonstrated more than 40 years ago that injection of Ca2+ in one cell of a coupled cell-pair inhibits electrical coupling between cells (Rose and Loewenstein, 1975). The closing of GJs is considered to be protective, because it will isolate the cell displaying a pathologic process, thereby avoiding GJ-mediated bystander effects on healthy neighbors. The [Ca2+]i at which GJs close is rather variable between different studies and ranges from 300 nM (Lazrak and Peracchia, 1993; Crow et al., 1994; Lurtz and Louis, 2007) to several micrometers (Rose and Loewenstein, 1975; Spray et al., 1982). This may result from differences between the involved connexins or differences between cell types; it may also be linked to the fact that global rather than microdomain [Ca2+]i was measured. It furthermore appears that Ca2+ entry is more effective in closing GJs compared with Ca2+ store release (Lazrak et al., 1994). Dakin et al. (2005), making use of a photoactivated fluorescent probe, elegantly demonstrated that capacitative Ca2+ entry following store release was particularly potent in inhibiting junctional coupling (Dakin et al., 2005). Although Ca2+-inhibition of GJs has been reported to be fast (Lazrak et al., 1994), others have reported slower kinetics sometimes in the order of minutes (Churchill et al., 2001). Gap junction closure by [Ca2+]i elevation is thought to be mediated by Ca2+-calmodulin signaling (reviewed in Peracchia, 2004; Lurtz and Louis, 2007). Cx32 has (at least) two calmodulin interaction sites located on the NT and CT (Torok et al., 1997), whereas Cx43 has one on its CL (Zhou et al., 2007) (Fig. 1). Ca2+-triggered GJ closure can be prevented by the calmodulin inhibitor W7 (Peracchia, 1987) and by a peptide that mimics the calmodulin interaction site on the CL of Cx43 (Zhou et al., 2007) (Lys-136–Ser-158 indicated as the “CaM” domain in Fig. 1). Peracchia and coworkers (Peracchia et al., 2000; Peracchia, 2004) suggested that Ca2+-calmodulin closure of Cx32 channels is mediated by a cork-plug model that involves channel pore obstruction by one of the calmodulin spherical lobes. Calmodulin may additionally act indirectly via calmodulin-dependent kinases below in section II.E.5.b). Finally, as remarked for pH effects, there are exceptional findings as well, in that [Ca2+]i elevation or Ca2+-linked signaling has been reported to increase GJ coupling in heart as well as brain (Alev et al., 2008; De Pina-Benabou et al., 2001; Delage and Délèze, 1998; Siu et al., 2016). Clearly, [Ca2+]i modulation of GJs may occur via multiple intermediate signaling steps, including various kinases; as a consequence, the kinetics can be slow and the outcome on channel function diverse.
b. Intracellular Ca2+ concentration effects on hemichannels.
Although [Ca2+]i elevation mostly closes GJs, its effects on HCs link to activation of channel opening. For Cx32, De Vuyst et al. (2006) reported, based on [Ca2+]i measurements and dye uptake or ATP release studies, that [Ca2+]i elevation triggers HC opening (De Vuyst et al., 2006); similar findings were reported in a follow up study performed on Cx43 (De Vuyst et al., 2009). In these studies, [Ca2+]i was increased by a Ca2+ ionophore, triggering Ca2+ entry, or by photoactivation of caged (inactive) IP3 that triggers ER Ca2+ release. The [Ca2+]i response curve was peculiar in that elevations up to 500 nM promoted HC opening, whereas [Ca2+]i elevation above 500 nM resulted in a gradual disappearance of the promotive effect, giving a convex-up bell-shaped response curve with a peak at 500 nM and closure at 1 µM [Ca2+]i (Fig. 2B). The closure at 1 µM [Ca2+]i possibly acts as a brake to prevent excessive HC opening; a similar brake—albeit mediated by ATP—exists for Panx1 channels where ATP released via the channel inhibits the channel (Qiu and Dahl, 2009). The bell shape of HC responses to [Ca2+]i elevation was also observed at the single channel level in HeLa cells overexpressing Cx43 (Wang et al., 2012a; Bol et al., 2016). In these cells [Ca2+]i modulated the HC opening activity triggered by voltage steps to positive voltages (+40 mV and higher) but it did not trigger HC opening by itself; imposing [Ca2+]i elevations in the absence of the electrical trigger were ineffective in opening HCs. [Ca2+]i however had another interesting effect in that it lowered the voltage activation threshold for HC opening; even small [Ca2+]i elevations from 50 to 200 nM lowered the voltage threshold by ∼15 mV (Wang et al., 2012a). Note that [Ca2+]e also affects the voltage activation of HC opening but here the activation threshold is lowered by a decrease in [Ca2+]e (Contreras et al., 2003). The HC enhancing effect of moderate (≤500 nM) [Ca2+]i elevation has not only been observed in Cx43 expressing HeLa cells but has also been reported for acutely isolated ventricular cardiomyocytes (Wang et al., 2012a). Recent evidence obtained in mouse astrocytes furthermore demonstrates HC opening in response to [Ca2+]i elevation with the membrane potential clamped at −70 mV, i.e., without associated electrical stimulation (Meunier et al., 2017); similar observations are available for mouse and pig ventricular cardiomyocytes (unpublished data). The mechanism of the [Ca2+]i enhancing effect on HC opening activity has been linked to calmodulin, calmodulin-dependent kinase II, and several other factors; calmodulin-activation was particularly robust and was also effective when it was activated in a Ca2+-independent manner by a Ca2+-like peptide called CALP (De Vuyst et al., 2009). The disappearance of enhanced HC function at higher, above 500 nM [Ca2+]i elevation (Fig. 2B) was shown to depend on actomyosin contractility that acts to disrupt intramolecular/intersubunit CT-CL interaction (Ponsaerts et al., 2010; reviewed in Ponsaerts et al., 2012). In line with this, CT-CL disruption by high (1 µM) [Ca2+]i was largely prevented by the myosin II ATPase inhibitor blebbistatin (Ponsaerts et al., 2008). The location of the actomyosin interaction site on the Cx43 CT is not known but the ZO-1 site has been excluded; involvement of the Drebrin site (see Fig. 1; Ambrosi et al., 2016), which like ZO-1 links to F-actin, still needs to be verified. Collectively, this work combined with the observations of Wang et al. (2013c), demonstrated that CT interaction with the CL is necessary for Cx43 HCs to become available for opening when a trigger is present; in the absence of CT-CL interaction, HCs are unresponsive to triggers and remain in the closed state (Fig. 2A) (Wang et al., 2013a,c). CT-CL interaction involves the CT9 domain (Fig. 1) that interacts with the CL-located L2 domain, whereby the negatively charged Asp-378 and Asp-379 residues on the CT play a crucial role (D’Hondt et al., 2013). Interestingly, supplying CT9 peptide prevented the HC closure at high (1 µM) [Ca2+]i and thus removes the HC brake, most likely by directly binding to the L2 domain on the CL thereby acting as a CT substitute (Bol et al., 2016). Thus, in HCs, at least those composed of Cx43, CT-CL interaction is necessary for HCs to open, whereas the very same interaction closes the GJs, possibly by a particle-receptor scenario as explained previously (Fig. 2A). Data of Schalper et al. (2008) with fibroblast growth factor (FGF)-1 stimulation are in line with these observations; these authors reported that FGF-1 triggered Cx43 HC opening was dependent on [Ca2+]i and lost upon CT truncation at position 257 (Schalper et al., 2008). It needs to be added that CT-truncated Cx43 (at Lys-258) still displays HC opening with low [Ca2+]e stimulation (Kozoriz et al., 2010), indicating that the CT-CL interaction hypothesis might only be valid for triggers that come from the intracellular side such as Vm steps to positive voltages or increases in [Ca2+]i. Of note, exposure of cells to low [Ca2+]e may provoke [Ca2+]i changes that act as the actual trigger of HC opening (De Vuyst et al., 2006). Besides effects of [Ca2+]i elevation on HC gating, Ca2+ may influence connexin trafficking as well: ER Ca2+ release, triggered by oxidative stress, promoted Cx43 HC opening by increasing the cell surface HC pool as determined from biotinylation experiments (Riquelme and Jiang, 2013).
Modulation of HCs and GJs in opposite directions, as observed in the context of CT-CL interaction (Fig. 2A), are not uncommon and have been reported for several conditions and signals. Also [Ca2+]i distinctly influences the two channel types, because it promotes HC opening (at least when [Ca2+]i is below 500 nM; see Fig. 2B) while it is well documented to inhibit GJs (exceptions exist here as well, as referred to earlier). Several other modulatory factors have opposite effects, most notably proinflammatory cytokines like tumor necrosis factor-α (TNF-α) and interleukin 1-β (IL1-β), or bacterial lipopolysaccharide (LPS), which all promote HC opening but close GJs (De Vuyst et al., 2007; Retamal et al., 2007a; Orellana et al., 2009); metabolic inhibition has a similar effect (Contreras et al., 2002). Along the same line, arachidonic acid inhibits GJs but promotes HC opening (Contreras et al., 2002; De Vuyst et al., 2007, 2009). By contrast, acidification inhibits both GJs and HCs (Spray et al., 1981; Trexler et al., 1999), a remarkable observation because low pHi closure of GJs is supposed to be mediated by CT-CL interaction while this very same interaction is expected to promote HC opening. However, Wang et al. (2012a) reported that Cx43 HC closure at low pHi disappears under conditions of strong cytoplasmic Ca2+ buffering with 1,2-bis(o-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid (commonly known as BAPTA), indicating that low pHi closure depends on [Ca2+]i elevation (Wang et al., 2012a). Another possibility is that acidification has more direct effects on Cx43 that do not depend on CT-CL interaction.
6. Other Gating Effects.
Several other HC opening triggers have been reported, but not all of them have been characterized at the electrophysiological level. Retamal et al. (2007b) reported that Cx43 HCs open in response to intracellularly applied reducing conditions making use of glutathione or dithiothreitol (Retamal et al., 2007b), an effect that was exercised at the level of channel gating. The latter experiments were done under normoxic conditions; under ischemic conditions (mimicked by metabolic inhibitors) astrocytic HCs also opened, but this was linked to an increase of the plasma membrane HC pool. The resulting increase of HC function was mediated by S-nitrosylation, possibly by NO acting at CT-located Cys residues, and inhibited by reducing agents (Retamal et al., 2006; reviewed in Sáez et al., 2005). Interestingly, NO is an HC permeant that may also activate HC opening (Retamal et al., 2006; Figueroa et al., 2013), as is the case for Ca2+ [Ca2+ permeates as well as activates channel opening by elevating [Ca2+]i; NO can freely diffuse over the plasma membrane but when open HCs are available, they may facilitate NO diffusion (Figueroa et al., 2013)]. Work from the Lal group demonstrated Cx32 and Cx43 HC opening in response to H2O2-induced generation of reactive oxygen species (ROS) that appeared to involve Vm changes (Ramachandran et al., 2007). Carbon monoxide (CO) inhibits Cx43- and Cx46-based HCs independently of CT Cys residues, and the effect disappears by application of extracellularly acting reducing agents (León-Paravic et al., 2014). Effects of phosphorylation on gating of GJs/HCs are addressed section II.F.
F. Connexin Posttranslational Modifications
Many different connexin posttranslational modifications have been reported, including phosphorylation, ubiquitination, sumoylation, S-nitrosylation, palmitoylation, hydroxylation, acetylation, methylation, and γ-carboxyglutamation. However, the evidence supporting their prevalence and biologic significance varies considerably. Evidence that can support the existence and importance of posttranslation modifications includes detection of the modification using multiple analytical techniques, finding the modification in cells and tissues and observation of a biologic effect if that posttranslational modification is prevented. For example, detection of a posttranslational modification via direct incorporation of a radioactive element that is part of the posttranslational modification, detection via mass spectrometry, and/or detection via an antibody to the modification in connexin purified from cells would be convincing evidence that a posttranslational modification can occur. However, detection via a single method can be potentially misleading for several reasons. For example, amino acid modifications can be introduced during protein purification steps and then they are easily detected via high sensitivity mass spectrometers currently in use today. In vitro phosphorylation of a connexin, often just the CT region combined with a purified kinase, can indicate a phosphorylation event is possible, but without evidence that it happens in cells, the result by itself is not particularly important. Artificial introduction of a modification can also occur via overexpression of the substrate or the enzyme that catalyzes the modification in cells. Because all chemical reactions even within cells depend on the concentrations of the reactants, they may be possible under artificial conditions but not occur significantly in vivo. Therefore, multiple lines of evidence supporting the modification including its presence in vivo gives one more confidence that a modification was not artificially introduced. Furthermore, if modulation of the modification in vivo changes the biologic properties of GJ communication or other connexin characteristics (e.g., changes in interacting proteins or connexin localization) that would be good evidence of biologic relevance. For example, good evidence that a posttranslational modification is biologically meaningful might be derived from detection of the posttranslational modification by multiple methods in vivo (e.g., immunoprecipitation and detection via mass spectrometry or isotope incorporation and detection in cells via an antibody to the modification). Biologic relevance could be shown via modulation of the extent of modification through application of a targeted stimulus that elicits a physiologic change in connexin properties and elimination of the modulatory effect of the phosphorylation event when the specific connexin amino acid is mutated so it cannot be modified.
1. Phosphorylation.
The best documented connexin posttranslational modification is phosphorylation. Connexin phosphorylation has been reported to affect connexin half-life, protein trafficking, incorporation into a GJ, GJ size, channel gating, and protein turnover (for recent reviews, see Pogoda et al., 2016; Solan and Lampe, 2016). Furthermore, wounding, ischemia, and other tissue insults have been shown to change connexin phosphorylation. Knock-in mice where three Ser to Ala or Ser to Glu point mutations were introduced in the gene for Cx43 at sites phosphorylated by casein kinase 1 (CK1) were highly susceptible or resistant to inducible arrhythmias, respectively (Remo et al., 2011). Knock-in mice expressing Cx43 with mitogen-activated protein kinase (MAPK) phosphorylation sites mutated to Ala showed reduced proliferation during arteriole injury and reduced neointima formation (Johnstone et al., 2012). Knock-in mice with Cx43 containing a Ser to Ala mutation at a protein kinase C (PKC) site were unresponsive to sphingosine-1-phosphate cardioprotective effects upon ischemia reperfusion injury (Morel et al., 2016). These results combined with those showing that these residues are phosphorylated in vivo essentially prove that connexin phosphorylation plays key roles in the physiologic response to injury.
Many connexins isolated from cellular preparations have been shown to incorporate radioactive phosphorous; demonstrate phosphatase-dependent shifts in mobility in SDS-PAGE; exhibit charge/mass ratios via mass spectrometry that are consistent with phosphorylation; and/or yield phospho-Ser, -Thr, and/or -Tyr in amino acid and peptide analyses. However, as for other posttranslational modifications, the extent of the evidence varies by connexin. There is at least one report of phosphorylation (discussed below) of most of the connexin α group members, including Cx37 (Morel et al., 2010), Cx40, Cx43, Cx46, and Cx50 and for Cx36, Cx45, and Cx47 (May et al., 2013), that belong to the γ group (α,β or γ group assignations; see Cruciani and Mikalsen, 2006). Phosphorylation of other connexin group members, including Cx29 (Wiśniewski et al., 2010), Cx31 (Diestel et al., 2004), Cx31.9 (Nielsen and Kumar, 2003), and Cx32 (discussed below), has also been reported, but little biologic data supporting their biologic relevance exist. At least two connexin proteins (Cx26 and Cx33) have been reported to be not phosphorylated. In the next several paragraphs, we discuss the connexins for which significant data exist.
a. Cx43 phosphorylation.
The most well-studied and characterized phosphoconnexin is Cx43. At least 19 of the 26 Ser and 4 of the 6 Tyr in the CT region of Cx43 have been identified as kinase substrates, and there has been some progress in the characterization of the network of kinases that phosphorylate Cx43 (phosphorylation sites are illustrated in Fig. 1 and major effects are summarized in Table 1). Of course, the level of evidence supporting these sites and the kinases involved varies considerably. Sites with high confidence include those phosphorylated by MAPK family members, PKC (especially δ and ε), Src, Akt, and CK1. MAPK3/MAPK1 (ERK1/2) are known to phosphorylate Ser-255, Ser-279, and Ser-282 in vitro (Warn-Cramer et al., 1996); these sites are phosphorylated in cells when MAPK is activated and GJ channel properties are modulated in a manner that is abrogated when the sites are mutated to Ala (Warn-Cramer et al., 1998). MAPK7 can also phosphorylate Ser-255 (Cameron et al., 2003), and Ser-262 has been identified as a substrate for CDK1 (Kanemitsu et al., 1998). Phosphospecific antibodies have been created for all of these sites and they react with Cx43 in Western blots and via immunofluorescence in the expected manner when MAPK is activated or inhibited. Although the exact sites and stoichiometry of the phosphorylation events are not clear, in vitro phosphorylation of Cx43 with purified CK1 yields phosphorylation at Ser-325, Ser-328, and Ser-330, and CK1 activation increases their phosphorylation, whereas inhibition decreases phosphorylation (Table 1). Assaying with a phosphospecific antibody for these sites reports decreased signal during hypoxia, and mutation of these sites affect the development of cardiac arrhythmia (Remo et al., 2011). Tyr-247 and Tyr-265 of Cx43 are known Src substrates and Tyr-301 and Tyr-313 were identified as phosphorylated in general MS/MS-based screens for phosphotyrosine-containing proteins (Ballif et al., 2008; Bonnette et al., 2010). A recent report indicates that Tyk2 can phosphorylate the same Tyr residues as Src (Li et al., 2016). The last 19 amino acids of Cx43 contain three sets of double Ser (Ser-364/Ser-365, Ser-368/Ser-369, Ser-372/Ser-373) that all have been reported to be phosphoacceptors (e.g., Yogo et al., 2002). Ser-368 is a very well-documented substrate for classic PKCs, and early creation of a phosphospecific antibody for this site has led to dozens of reports of Ser-368 modulation in response to various cellular treatments and conditions. Akt phosphorylates Cx43 primarily at Ser-373 and secondarily at Ser-369 in vitro (Park et al., 2007), and the binding of a phosphospecific antibody for phosphorylated Ser-373 is decreased when Akt is inhibited in cells (Dunn and Lampe, 2014). Ser-364 (TenBroek et al., 2001) and Ser-365 (Solan et al., 2007) are also known to be phosphorylated particularly under conditions where PKA is activated, but the kinase(s) that actually phosphorylate these residues is less clear as it is a poor substrate for direct phosphorylation by PKA (TenBroek et al., 2001; Shah et al., 2002).
Residues, kinases, and effects of Cx43 phosphorylation
There are other documented Cx43 phosphorylation sites for which the kinase responsible is less clear. Ser-262 has been reported to be phosphorylated in response to PKC activation (Doble et al., 2004; Srisakuldee et al., 2014), but it was not a major substrate for purified classic PKCs (Sáez et al., 1997; Lampe et al., 2000), raising the question of whether PKC activation could cause activation of other kinases, potentially proline-directed kinases, because there is a Pro C terminus to Ser-262 fitting the consensus sequence and it is a known substrate for a proline-directed kinase (Kanemitsu et al., 1998). A similar issue was observed for Src activation, which subsequently leads to MAPK and PKC activation and phosphorylation at Ser-255, Ser-262, Ser-279, Ser-282, Ser-368, Tyr-247, and Tyr-265 (Solan and Lampe, 2008), potentially leading to confusion as to which sites and kinases actually are responsible for specific effects (Lin et al., 2001; Mitra et al., 2012). There are other reported sites for which we have less information, like Ser-296, Ser-297, Ser-306, and Ser-314 (Procida et al., 2009), which need to be further investigated. Nonetheless, we know much more about Cx43 regulation via phosphorylation compared with the other connexins.
b. Cx32 phosphorylation.
Cx32 was shown to be phosphorylated by metabolic labeling of hepatocytes, and phosphorylation levels were shown to increase with 8-bromo-cAMP (Sáez et al., 1986), forskolin, or phorbol ester treatment (Sáez et al., 1990). Ser-233 appeared to be a substrate for both PKA and PKC, and its phosphorylation level increased when the kinases were activated in cultured cells (Sáez et al., 1990). Epidermal growth factor receptor tyrosine kinase may also be involved in phosphorylation of Cx32 on tyrosine residues (Díez et al., 1998). However, the physiologic effects of all of the Cx32 phosphorylation events need to be studied further to better understand their biologic implications.
c. Cx36 phosphorylation.
There has been a great deal of interest in the regulation of Cx36, because it has been found to be expressed in neurons as well as in dendrites of AII amacrine cells of the retina. Cx36 was shown to be a phosphoprotein by metabolic labeling and mass spectrometry (Urschel et al., 2006). Phosphospecific antibodies to PKA substrates Ser-110 and Ser-276 in Bass (Ser-293 in mammals) showed that large GJs in the inner plexiform layer showed higher levels of phosphorylation in dark-adapted and reduced levels in light-adapted retina (Kothmann et al., 2007). Dopamine-driven, PKA-dependent uncoupling of the AII amacrine cell network occurs via PKA activation of protein phosphatase 2A and subsequent dephosphorylation of Cx36 (Kothmann et al., 2009).
d. Cx40 phosphorylation.
Cx40 was shown to be phosphorylated via metabolic labeling (Traub et al., 1994). 8-Br-cAMP addition to Cx40-transfected cells resulted in a SDS-PAGE mobility shift in Cx40 and increased macroscopic GJ conductance between cell pairs (van Rijen et al., 2000). LPS or hypoxia and reoxygenation treatment of microvascular endothelial cells led to a reduction in Cx40 phosphorylation and electrical coupling between cells that was dependent on PKA (Bolon et al., 2008).
e. Cx45 phosphorylation.
Cx45 has been convincingly shown to be phosphorylated via metabolic incorporation and phosphatase-dependent SDS-PAGE mobility shifts at primarily serine residues (Butterweck et al., 1994; Darrow et al., 1995), with less phosphotyrosine and phosphothreonine also being reported (Hertlein et al., 1998). Serines present within residues 374–393 were responsible for 89% of the phosphorylation and, in particular, Ser-381, -382, -384, and/or -385 were found to be important in regulating Cx45 stability, because mutation of different combinations of these reduced the half-life of the mutant version to 50% of wild-type Cx45 (Hertlein et al., 1998). Treatment of cells expressing Cx45 with 8-Br-cAMP or pervanadate increased Cx45 phosphorylation levels and decreased GJ conduction (van Veen et al., 2000).
f. Cx46 and Cx50 phosphorylation.
These connexins are usually coexpressed in tissues, and several reports examined the phosphorylation status of both proteins so they are covered here together. Both migration shifts (Pelletier et al., 2015) and immunoprecipitation followed by mass spectrometry (Lin et al., 2004) have shown Cx46 and Cx50 to be phosphorylated at 9 and 18 serine residues, respectively (Wang and Schey, 2009). PKA phosphorylated Ser-395 of Cx50 and mutation of this residue to alanine attenuated increases in dye coupling and uptake caused by PLA activation (Liu et al., 2011). Phosphorylation of Cx50 can accelerate turnover (Yin et al., 2000) and susceptibility to caspase-3-like protease cleavage (Yin et al., 2001).
2. Connexin Ubiquitination.
Protein ubiquitination controls many aspects of cellular function by targeting substrates for degradation via the proteasome and through ubiquitin-dependent protein-protein interactions. There has been at least one report that Cx26 (Xiao et al., 2014), Cx32 (Kelly et al., 2007), and Cx40 (Gemel et al., 2014b) can be ubiquitinated, and there are many reports for Cx43. Furthermore, inhibition of the proteasome dramatically increases the levels of Cx40 (Gemel et al., 2014b) and Cx43 present within GJs (VanSlyke and Musil, 2005; Dunn and Lampe, 2014). However, ubiquitin signaling is complex and there are many inconsistencies in the literature, potentially due to the variety of cellular systems under study that complicate our understanding of its role in connexin biology. For example, different reports indicate that Cx43 is mono-ubiquitinated (Leithe and Rivedal, 2004; Girão et al., 2009) or it is poly-ubiquitinated (Laing and Beyer, 1995; Laing et al., 1997; Ribeiro-Rodrigues et al., 2014; Martins-Marques et al., 2015b) and that it can preferentially be degraded by the proteasome or the lysosome. Given that the biologic consequences of mono- versus poly-ubiquitination are potentially dramatically different (i.e., resulting in putative proteasomal degradation or activity regulation), these inconsistencies need some resolution. Most of these studies used cell lines where the ubiquitin is epitope or 6xHis tagged and overexpressed. Also, most used a Cx43 antibody to immunoprecipitate Cx43 and then cut the blot in two and probed the top half with antiubiquitin to show a ladder or smear of signal and the lower half with anti-Cx43. The problem with this approach is that the levels and exposures of the two halves are not comparable and any protein that coprecipitates with Cx43 might be the actual ubiquitin acceptor rather than Cx43. In studies where estimates of the level of the putative ubiquitinated Cx43 isoform were made, it appeared to be much less than 1% of the total Cx43 comigrated with what might be an ubiquitinated isoform under conditions that should favor its accumulation (Girão et al., 2009; Chen et al., 2012b). Furthermore, conversion of all Lys in Cx43 to Arg to eliminate ubiquitination did not affect the accumulation of Cx43 in GJs in response to proteasomal inhibitors (Dunn et al., 2012) or during ER-localized degradation (Su et al., 2010). In fact, two other Cx43-interacting proteins that are ubiquitin substrates or ubiquitin associated, Akt (Dunn and Lampe, 2014) and CIP75 (Su et al., 2010), were found to regulate these processes in ubiquitin-dependent processes. Therefore, caution should be exercised when ubiquitin-related data are interpreted. On the other hand, a fusion protein of Cx43 and ubiquitin had a shorter half-life than wild-type Cx43 (Catarino et al., 2011). Clearly the ubiquitin system regulates connexin distribution. However, given its low levels and lack of a role in GJ size regulation and ER-associated degradation, further research needs to be performed to define the biologic function of Cx43 and connexin ubiquitination in general. One possible explanation consistent with at least some of these results is that ubiquitination of Cx43 does occur but only under specific situations such as autophagy (further discussed section II.F.4 below).
3. Other Potential Connexin Posttranslational Modifications.
The isoelectric point of Cx43 is lower than predicted from its sequence in a manner thought to be independent of phosphorylation (Stockert et al., 1999), suggesting other modifications might be possible. A handful of reports indicate that connexins (e.g., Cx26 and Cx43) are also posttranslationally modified by S-nitrosylation (Retamal et al., 2006), SUMOylation (Kjenseth et al., 2012), acetylation (Colussi et al., 2011), palmitoylation, hydroxylation, methylation, and γ-carboxyglutamination. Methylation, acetylation, hydroxylation, palmitoylation, and γ-carboxyglutamination of Cx26 were reported as possible posttranslational modifications in an MS-based study of purified Cx26 (Locke et al., 2009). Although intriguing, as pointed out previously (Chen et al., 2013), this MS study used a limited search strategy based on accurate masses expected for Cx26 fragments rather than MS/MS analysis using broad search strategies, and this could have easily generated false discovery. Furthermore, the highly sensitive mass spectrometers in use today can generate false-positive posttranslational modification detection at many steps from protein purification (e.g., silver staining of gels) to MS data peak assignment (Larsen et al., 2006; Kim et al., 2016a). Therefore, confirmation via other techniques and studies will be necessary for all of these modifications, and if confirmed, understanding their biologic effect will become important.
4. Regulation of the Connexin Life Cycle via Posttranslational Modifications.
a. Connexin protein half-life.
Although GJs appear as complex structures and can be difficult to disrupt during biochemical isolation, most connexin proteins have been shown to have short half-lives (1–5 hours), both in vivo and in cell culture (Crow et al., 1990; Musil et al., 1990; Laird et al., 1991; Lampe, 1994; Darrow et al., 1995; Beardslee et al., 1998; Hertlein et al., 1998; Lauf et al., 2002; Laird, 2006, 2010). However, the half-life of Cx30 (Kelly et al., 2015) and some isoforms of Cx50 (Jiang and Goodenough, 1998; Berthoud et al., 1999) appear to be much longer and more similar to typical integral membrane proteins (Chu and Doyle, 1985; Hare and Taylor, 1991). Why a cell would turnover connexin proteins so rapidly is not obvious. A short half-life with extensive regulation could allow the cell to control exquisitely processes dependent on connexin-binding protein interactions, HC function, or GJ communication. At least one region of Cx43 can control turnover, because it contains two putative Tyr-based sorting signals [Yxxφ; where φ = hydrophobic (Fong et al., 2013)], including the key sequence Y286KLV (see Fig. 1), which upon Val to Asp mutation displays a threefold increased protein half-life (Thomas et al., 2003). As integral membrane proteins, connexins are synthesized in the ER but they lack a canonical membrane signal sequence and at least Cx43 does not oligomerize into a hexameric HC until reaching the trans-Golgi network (Musil and Goodenough, 1993). This delay in multimerization is hypothesized to provide a quality control step, because GJ assembly can be downregulated through endoplasmic reticulum-associated degradation during conditions of cellular stress (VanSlyke and Musil, 2002; Su et al., 2010).
b. Regulation of connexin export to the plasma membrane and gap junction assembly.
Live cell and other imaging techniques show multiple regulatory aspects of Cx26, Cx32, and Cx43 transport to the plasma membrane, including microtubule-based vesicle transport (Martin et al., 2000; Johnson et al., 2002; Lauf et al., 2002; Shaw et al., 2007). Imaging studies have also shown that Cx43 can move from the plasma membrane into the periphery of a larger plaque; thus plaques grow by adding channels to the outside, and the oldest proteins are found in the center of the plaque where they get selectively turned over (Gaietta et al., 2002). Live cell imaging techniques also show Cx43 is highly motile with dynamic interactions with the cytoskeleton and events at GJs (e.g., Jordan et al., 1999, 2001; Martin et al., 2001; Lauf et al., 2002; Murray et al., 2004; Fiorini et al., 2008; Solan and Lampe, 2016). Phosphorylation of Cx43 can occur within 15 minutes of synthesis (Crow et al., 1990), and several kinases have been reported to regulate the assembly of GJs. When examined via Western immunoblot, many connexins demonstrate a phosphorylation-dependent reduction in SDS-PAGE mobility. Cx43 can often show three prominent bands sometimes labeled as P0, P1, and P2. Antibodies to Cx43 phosphorylated at Ser-365 show the P1 and P2 phosphoisoforms, which occur during the transition from the cytoplasm to the plasma membrane (Solan et al., 2007; Sosinsky et al., 2007). Furthermore, Ser-365 phosphorylation plays a “gatekeeper” role by preventing downregulation of GJ communication by subsequent Cx43 phosphorylation at Ser-368 (Solan et al., 2007). Several studies show that activation of cAMP-dependent protein kinase (PKA) can stimulate Cx43 trafficking to the plasma membrane (Atkinson et al., 1995; Burghardt et al., 1995), resulting in enhanced GJ assembly (see Table 1) and increased phosphorylation at many of the last six serine residues in Cx43 (TenBroek et al., 2001; Yogo et al., 2002,2006; Solan et al., 2007). CK1 (Table 1) phosphorylates Cx43 on some combination of residues Ser-325, Ser-328, and/or Ser-330 during the transition of Cx43 from the plasma membrane into the GJ (Cooper and Lampe, 2002), and a phosphospecific antibody to this region recognizes only Cx43 present in the GJ and the P2 form of isoform Cx43 (Cooper and Lampe, 2002). Akt can phosphorylate Cx43 primarily at Ser-373 (see Table 1) and this step allows GJs to grow in size via a reduced interaction with ZO-1 (Dunn and Lampe, 2014), which eliminates the well-documented ability of ZO-1 to restrict GJ size (Hunter et al., 2005; Rhett et al., 2011). Although Cx43 GJ assembly is clearly regulated by a series of kinases that can fine tune GJ communication, our understanding of how GJ assembly is regulated for the other connexins is much less well understood.
c. Regulation of gap junction turnover.
Although we understand many aspects of GJ assembly, particularly for Cx43, and connexin protein turnover can easily be examined using pulse-chase methods, several aspects of GJ turnover remain unclear. For example, live cell imaging studies show that a single GJ can be relatively stable for hours but a neighboring one can be rapidly dispersed (e.g., Solan and Lampe, 2016). Proteasomal inhibitors can stabilize Cx43-containing GJs and make them larger, but they usually only have a mild effect on the total level of Cx43 (Lampe et al., 1998; Qin et al., 2003). However, lysosomal inhibitors can increase Cx43 protein levels, but the protein appears to build up in cytoplasmic membranes (Qin et al., 2003; Berthoud et al., 2004). Internalized annular junctions (connexosomes) (Severs et al., 1989; Jordan et al., 2001; Laird, 2006; Leithe et al., 2006; Piehl et al., 2007; Fong et al., 2012; Johnson et al., 2013; Nickel et al., 2013) are apparent in some cell types and in cells undergoing autophagy, where they colocalize with the clathrin-adapter proteins disabled2 (Piehl et al., 2007), Atg14 and 9 (Bejarano et al., 2014), and the autophagosome membrane protein LC3 (Hesketh et al., 2010; Fong et al., 2012). Autophagy and degradation of multiple connexin isoforms have been linked (Iyyathurai et al., 2016). However, it is unclear whether this mechanism might account for most/all of GJ turnover or whether it occurs only in specialized circumstances such as nutrient deprivation. Part of the issue is that different treatments that affect clathrin-mediated and other internalization processes have only partial effects, making firm conclusions more difficult. Clearly more research is needed to understand whether multiple mechanisms for GJ turnover exist in vivo.
Although our knowledge of the mechanisms involved in GJ turnover is incomplete, we know that epidermal growth factor, 12-O-tetradecanoylphorbol acetate, Src activation, wounding, and extracellular ATP (Kanemitsu and Lau, 1993; Ruch et al., 2001; Schwiebert and Zsembery, 2003; Rivedal and Leithe, 2005; Chang et al., 2008; Dunn and Lampe, 2014) lead to Cx43 phosphorylation and loss of GJs. PKC (Lampe, 1994; Solan et al., 2003; Richards et al., 2004), MAPK (Johnstone et al., 2012), Src (Solan and Lampe, 2008), and Akt (Dunn et al., 2012) probably play at least some role in the regulation of Cx43 turnover. A kinase program that spatiotemporally activates Cx43 GJ turnover via sequential phosphorylation by Akt, MAPK, Src, and PKC in response to growth factors, wounding, and other stimuli has been proposed (Solan and Lampe, 2016). In this model, an Akt-mediated transient increase in GJ size depletes the nonjunctional Cx43 by rapid incorporation into a GJ, and the resulting larger GJ potentially reduces the energetics of annular junction formation by enabling membrane curvature during internalization. Src can clearly play a role, because its inhibition via PP2 treatment blocks growth factor-induced GJ turnover (Spinella et al., 2003; Gilleron et al., 2008). For many years it has been known that v-Src activity can downregulate GJ communication (Atkinson et al., 1981; Azarnia et al., 1988; Menko and Boettiger, 1988) coincident with an increase in tyrosine phosphorylation on Cx43 (Crow et al., 1990; Swenson et al., 1990). In LA25 cells that express active v-Src, Cx43 residues Tyr-247 (Src), Tyr-265 (Src), Ser-255 (MAPK), Ser-262 (MAPK), Ser-279/282 (MAPK), and Ser-368 (PKC) are all phosphorylated indicating co-activation of MAPK and PKC upon Src activation (Solan and Lampe, 2008). Phosphorylated Tyr-247 appeared to be preferentially present in larger GJ plaques (Solan and Lampe, 2014). Whether this distinct phosphoTyr-247 staining could potentially mark a portion of the GJ to facilitate interaction with components of the endocytic system is unknown. Src phosphorylation of Cx43 may trigger GJ endocytosis like it does with the NMDA receptor GluN3A (Chowdhury et al., 2013). There may also be three-way crosstalk regulation involved, because both ZO-1 and Src can bind to the CT region of Cx43 and they can bind to each other as well (Sorgen et al., 2004; Gilleron et al., 2008; Kieken et al., 2009), whereas Akt phosphorylation at Ser-373 inhibits ZO-1 binding to Cx43 (Dunn and Lampe, 2014). Furthermore, the GJ blocker glycyrrhetinic acid causes GJs to adopt a looser packing arrangement (Goldberg et al., 1996) in a process that involves Src binding (Chung et al., 2007) and that leads to disruption of Cx43-ZO-1 interaction (Gilleron et al., 2008).
d. Phosphorylation effects on connexin channel gating.
Phosphorylation effects on connexin channel function have been amply documented based on dye transfer studies making use of various channel-permeant fluorescent molecules. The data obtained with such approaches can be interpreted in several ways, because changes in dye transfer can be caused by alterations in the number of channels available, in the permeability properties of the channels, or in the channel gating characteristics. More fundamental insights into how phosphorylation alters GJ/HC function necessitates electrophysiological analysis at the single channel level, allowing the resolution of effects on open probability and single channel conductance, including main and residual conductance states. Only a few reports, reviewed in Moreno and Lau (2007), are available that have documented phosphorylation effects at a detailed unitary channel activity level. Kwak et al. (1995b) showed that cGMP-dependent Cx43 phosphorylation at Ser-259 (present in rat but not in human) reduced the GJ macroscopic conductance by decreasing the single channel main open state to a state with a lower conduction (Kwak et al., 1995b). Follow up work demonstrated distinct effects of phosphorylation by PKA, PKC, and PKG on different connexins (Cx26, Cx43, and Cx45) based on their biophysical single channel signature (Kwak et al., 1995a). Lampe et al. (2000) demonstrated that PKC phosphorylation at Ser-368 inhibits Cx43 GJs by shifting the unitary event activity in conductance histograms from 100 pS centered events, corresponding to the main open state, to 50–60 pS events that correspond to a residual subconductance state; this work was based on comparing channel event activities in Cx43 wild type with those in a Cx43-S368A phosphodead mutant (Lampe et al., 2000). However, PKC has also been reported to increase the open probability of Cx43 GJs (Kwak et al., 1995c; Kwak and Jongsma, 1996), making the net effect on electrical current less clear. Bao et al. (2007) demonstrated that PKC phosphorylation, in addition to the effects on gating, also alters the size selectivity as suggested by sucrose permeation experiments (Bao et al., 2007).
For v-Src, Cottrell et al. (2003) reported that GJ inhibition was not mediated by alterations in the unitary conductances; this work was based on comparing event activities in Cx43 wild type with those in cells expressing Cx43-Y247F,Y265F double mutant versions (Cottrell et al., 2003). Based on the fact that v-Src did not change the size of GJ plaques, suggesting that the number of GJ channels was not affected (Atkinson et al., 1986; Lin et al., 2001), the most probable effect of v-Src appeared to reside at the level of GJ channel open probability (Moreno and Lau, 2007). Further dye transfer studies by Cottrell et al. (2003) demonstrated that GJ channel permeability for Lucifer yellow and [2-(4-nitro-2,1,3-benzoxadiazol-7-yl)aminoethyl]trimethylammonium was also affected by v-Src phosphorylation at Tyr-247 and Tyr-265; the effects were stronger at the level of dye transfer than at the level of current inhibition, which looks similar to the PKC effects. Interestingly, CT truncation at residue 245, just two positions in NT direction relative to Tyr-247, removes GJ inhibition by v-Src (Zhou et al., 1999), making it possible that v-Src inhibition of GJs is the consequence of CT-CL particle-receptor (ball and chain) interaction (see section II.E.2); others have also linked v-Src to CT-CL interaction (Lau, 2005). Zhou et al. (1999) suggested that v-Src effects are indirect and mediated by MAPK effects at Ser-255, Ser-279, and Ser-282 (Warn-Cramer et al., 1998). In line with this, Cottrell et al. (2003) found that epidermal growth factor-activation of MAPK resulted in GJ inhibition that was not associated with alterations in the conductance states, as observed for v-Src; these authors furthermore demonstrated that MAPK-inhibition of GJs was absent in Cx43-S255A,279A,282A mutants. Of note, MAPK effects are fast (Kim et al., 1999) and have also been linked to CT-CL interaction (Harris, 2001).
Little is known about phosphorylation effects on the gating of channels composed of connexins other than Cx43. Sáez et al. (1986) reported that cAMP-stimulated phosphorylation of Cx32 induced a higher conductance state (Sáez et al., 1986). Similarly, van Rijen et al. (2000) reported that PKA activation, a well-known activator of GJ coupling downstream of cAMP, promoted Cx40 coupling by favoring a higher conductance state (van Rijen et al., 2000). Van Veen et al. (2000) demonstrated that pervanadate, which activates Tyr kinases, inhibited Cx45-based GJs by an effect that was concluded, based on exclusion of other possibilities, to decrease open probability (van Veen et al., 2000).
The same paucity of data applies to effects of phosphorylation at the level of gating of HCs. Several papers have suggested that MAPK family members may promote Cx43 HC opening (Schalper et al., 2008; De Vuyst et al., 2009; Avendano et al., 2015), but others have reported opposite effects (Kim et al., 1999; Riquelme et al., 2015). These results were based on indirect measures of HC function, mostly dye uptake or ATP release assays. Dye uptake studies also demonstrated that Akt/PKB, activated by metabolic inhibition, increased Cx43 HC function by increasing the HC pool in the plasma membrane as judged from biotinylation experiments (Salas et al., 2015). For PKC, all available data seem to point to inhibition of HC activity (Bao et al., 2004, 2007; De Vuyst et al., 2007); Hawat and Baroudi (2008) confirmed this at the electrophysiological macroscopic conductance level and further demonstrated that PKCε had more potent effects than PKCβ or PKCδ (Hawat and Baroudi, 2008). Clearly, more work is necessary here to resolve HC modulation by various kinases at the unitary current level and to determine how it affects gating and whether this relates to CT- or NT-linked conformational changes. Table 1 summarizes the effects of various kinases and target residues.
III. Pharmacological Modulation of Connexin Channels
The pharmacology of connexin channels is complex. First of all, specific inhibitors like those available for Na+, Ca2+, and K+ channels are not available. Second, most substances inhibit both GJs and HCs, except for some substances that are discussed below. Third, almost all inhibitors act in the micromolar range, not in the nanomolar range. Pharmacological modulation of GJs/HCs can, in principle, be achieved in various ways acting at distinct levels, including 1) connexin biosynthesis/expression, 2) connexin trafficking and HC assembly, 3) docking of HCs and formation of GJs, 4) formation of GJ plaques, 5) GJ disassembly by internalization and degradation, 6) the number of channels (by mechanisms related to 1 to 5), 7) channel gating, and 8) pore block or alterations in channel permeability when pore block is incomplete. Second messengers, posttranslational modifications, ions like Ca2+, H+, and membrane potential all act at least at one of these various levels, as described in the preceding sections. Several pharmacological agents may act at these different levels, either directly or indirectly by influencing intracellular signaling cascades (Table 2; reviewed in Dhein, 2004; Salameh and Dhein, 2005). For example, the fatty acid oleic acid inhibits GJ channels via PKC phosphorylation of Ser-368 (Huang et al., 2004). Lindane, a carcinogenic insecticide, inhibits GJs by MAPK activation that promotes the displacement of membrane-located Cx43 to the intracellular compartment (Mograbi et al., 2003). Tedisamil, a class III antiarrhythmic agent, promotes GJ conductance by acting on PKA (De Mello and Thormahlen, 1999).
Overview of effects of agents acting at various levels of the connexin life cycle
A. Connexin Channel Inhibitors
Several chemical classes of connexin channel inhibiting substances exist (reviewed in Salameh and Dhein, 2005; Bodendiek and Raman, 2010; Verselis and Srinivas, 2013). The best known and widely used substance in experimental work is carbenoxolone, a derivative of the licorice-derived glycyrrhetinic acid, which has better water solubility compared with 18-α-glycyrrhetinic acid and 18-β-glycyrrhetinic acid; carbenoxolone inhibits GJs with an IC50 in the range of ∼50 µM (see Table 1 in Spray et al., 2006). The mechanism of action of this family of licorice-based molecules on connexin channels is poorly defined. 18-α-Glycyrrhetinic acid has been reported to inhibit connexin synthesis (by decreasing transcription) and/or to promote connexin turnover at concentrations above 20 µM (Guo et al., 1999), whereas 18-β-glycyrrhetinic acid has been demonstrated to dephosphorylate Cx43 by type 1 or type 2A protein phosphatases (Guan et al., 1996). For carbenoxolone, several authors reported reduced Cx43 expression after prolonged (6–24 hours) carbenoxolone exposure (Herrero-González et al., 2009; Wang et al., 2009; Yulyana et al., 2013; Kim et al., 2017). Goldberg et al. (1996) reported no changes in connexin expression or location but alterations of HC packing in GJ plaques (Goldberg et al., 1996). Carbenoxolone is not a specific connexin channel blocker; for example, it is a commonly used drug in the treatment of gastrointestinal ulceration in Asian countries. Because of its steroid hormone backbone structure, it has intrinsic mineralocorticoid effects. Additionally, it has several other side effects (reviewed in Connors, 2012): it inhibits 11-β-hydroxysteroid dehydrogenase and thereby influences glucocorticoid synthesis (Monder et al., 1989; Sandeep et al., 2004), inhibits voltage-gated Ca2+ currents [IC50 ∼50 μM (Vessey et al., 2004)], has direct effects on GABA receptors (Connors, 2012), inhibits synaptic currents and action potentials (Rekling et al., 2000; Rouach et al., 2003; Tovar et al., 2009; Beaumont and Maccaferri, 2011), and acts as an anti-inflammatory agent (Amagaya et al., 1984; Inoue et al., 1989), which may all contribute to the observed effects beyond the effect on connexin channels as illustrated in, e.g., cardiac ischemia models (Haleagrahara et al., 2011). Carbenoxolone also inhibits pannexin channels with an IC50 ∼5 μM (Bruzzone et al., 2005), whereas higher concentrations (∼50 μM or higher) are necessary to inhibit GJs (see Table 1 in Spray et al., 2006); of note, connexin HCs are also inhibited by concentrations in the IC50 range of 5 µM. Also P2X7 receptors are inhibited by carbenoxolone with an IC50 in the range of 175 nM (Suadicani et al., 2006).
Other nonspecific connexin channel inhibitors include long-chain alcohols such as heptanol and octanol, which incorporate in the plasma membrane and act by altering membrane fluidity that somehow affects the channel (Bastiaanse et al., 1993). Another lipophilic agent is halothane, an inhalational general anesthetic that inhibits GJs at supra-anesthetic concentrations (Wentlandt et al., 2006)) in the order of 2 mM (Burt and Spray, 1989); it inhibits GJs by decreasing the open probability (He and Burt, 2000). Fatty acids like oleic acid, a mono-unsaturated fatty acid present in olive oil with various biologic effects, inhibits GJs indirectly via PKC phosphorylation, as mentioned above. Fatty acid amides like anandamide, an endogenous cannabinoid receptor agonist, also blocks GJs (Venance et al., 1995). The fatty acid arachidonic acid, a poly-unsaturated fatty acid in plasma membrane phospholipids involved in inflammatory signaling, takes a special place, because it inhibits GJs but promotes HC opening (Contreras et al., 2002; De Vuyst et al., 2007,2009). Its KD for GJ inhibition is ∼4 µM, and inhibition has been linked to decreased open probability (Schmilinsky-Fluri et al., 1997).
Fenamates like flufenamic acid, niflumic acid, and meclofenamic acid are nonsteroidal anti-inflammatory molecules that inhibit cyclo-oxygenase. They inhibit GJs with an IC50 in the order of 25–40 µM (Harks et al., 2001; Srinivas and Spray, 2003) and have been also demonstrated to block HCs (Gomes et al., 2005). Flufenamic acid acts by decreasing the open probability (Srinivas and Spray, 2003). Quinine and mefloquine are antimalarial drugs that display some specificity for inhibiting Cx36 and Cx50; mefloquine displays an IC50 of 300 nM and ∼1 µM for Cx36 and Cx50 (Cruikshank et al., 2004), i.e., very low concentrations compared with other connexin channel inhibitors, making it the most potent GJ inhibitor currently known. However, the downside is that the substance also has multiple side effects (see Verselis and Srinivas, 2013). Quinine has been shown to promote HC opening (Stout et al., 2002), similar to arachidonic acid. Derivatives of the triarylmethane cotrimoxazole, an antibiotic, also has connexin specificity targeted to Cx50; it acts with an IC50 of 1–2 µM and has a 10-fold higher specificity for Cx50 compared with other connexins as well as Na+ and K+ channels (Bodendiek et al., 2012; reviewed in Verselis and Srinivas, 2013).
Several other connexin channel-inhibiting compounds are best known for their other actions, including 2-amino ethoxydiphenyl borate, a blocker of IP3 receptors and store-operated Ca2+ entry; polyamines like spermine and spermidine acting on NMDA channels; and several other targets, 5-nitro-2-(3-phenyl-propylamino)benzoic acid, a chloride channel blocker; disodium 4,4′-diisothio cyanatostilbene-2,2′-disulfonate, an anion transport blocker; and certain triphenylmethanes, triphenylethanes, triarylmethanes, and cyclodextrins (Bodendiek and Raman, 2010).
B. Peptide Modulators of Connexin Channels
Improved specificity has been sought by making use of peptides that interact, directly or indirectly, with the connexin protein and thereby modulate, inhibit, or promote its channel function. Most of the peptides used are identical to a sequence on the connexin protein, e.g., Gap26 and Gap27, whereas others like AAP10 are unrelated to the connexin protein.
1. Mimetic Peptides of the Extracellular Loops.
a. GAP26 and GAP27 peptides.
Warner et al. (1995) were the first to report that peptides identical to sequences on the connexin EL domains (called “connexin mimetic peptides”) inhibited the synchrony of chick myoball contractions used as an indirect assay of GJ coupling (Warner et al., 1995). The starting point for this work was the hypothesis that supplying exogenous peptides identical to crucial domains on the ELs would interact with the ELs and thereby interfere with or prevent the docking of opposed HCs. By using their myoball approach, they identified several conserved motifs in the ELs that inhibited GJs when applied as synthetic peptides. For EL1, the crucial conserved connexin motifs were QPG and SHVR, whereas the conserved SRPTEK motif was crucial for EL2 (Warner et al., 1995). Peptides containing these motifs were, respectively, dubbed Gap26 (EL1) and Gap27 (EL2). For Gap26, the crucial motifs rather look like VCYD and FPISH for various connexins (e.g., Cx26, Cx32, Cx37, Cx40, and Cx43 and others) instead of QPG and SHVR (VCYD and FPISH motifs are illustrated in Fig. 1). The ability of SRPTEK-containing peptides to inhibit GJ coupling was enhanced by adding several amino acids from the putative membrane-spanning region. For example, Gap27 (Cx43 sequence SRPTEKTIFII in human and mouse) contains three amino acids that dig into the plasma membrane at TM4 (Fig. 1). Inhibition of GJs by Gap26/Gap27 is characterized by an IC50 of 20–30 µM (Chaytor et al., 1997) and the peptides need to be applied at 200–300 µM to obtain maximal effects. However, these data were not obtained directly by measuring GJ coupling but by indirect measurements consisting of phenylephrine-induced rhythmic activity of endothelium-denuded rabbit superior mesenteric artery rings. GJ block by Gap26/27 is never complete [∼95% complete for Gap26 (Desplantez et al., 2012b)]. Although Gap26 and Gap27 contain the conserved sequences VCYDXXFPISH and SRPTEK, respectively, they also contain amino acids that are less conserved. For example, the Gap27 sequence for Cx43 is SRPTEKTIFII (see Fig. 1; further referred to as 43Gap27), which is the same sequence as for Cx37 (hence its denomination as 37,43Gap27) but slightly different for Cx40: SRPTEKNVFIV (40Gap27). By contrast, the Gap26 sequence for Cx37 and Cx40 is the same: VCYDQAFPISHIR (37,40Gap26) but different for Cx43 (VCYDKSFPISHVR, Fig. 1). Experimental work with these peptides (or close derivatives) on vascular cells has demonstrated distinctive effects on Cx37, Cx40, and Cx43 (Kwak and Jongsma, 1999; Ujiie et al., 2003; Martin et al., 2005; Young et al., 2008; Billaud et al., 2011), suggesting they display some isoform, specificity.
The exact way of how Gap26 and Gap27 inhibit GJs is currently not known but several hypotheses exist (Evans and Boitano, 2001). First, the peptides may interact with the ELs of HCs, thereby preventing the interaction and docking of two complementary HCs to form a new GJ channel (Leybaert et al., 2003). As a result, fewer HCs will be incorporated into GJs (see section II.B). Because the connexin half-life is 1–5 hours, GJ inhibition would in this case be expected to occur on a similar time scale as hindered docking would decrease the number of GJ channels as a result of the ongoing GJ disassembly process. The kinetics of Gap26/Gap27 inhibition of Cx43-based GJs has an estimated half-time of ∼13 minutes based on electrophysiological measurements (Desplantez et al., 2012b). Indirect measurements based on Ca2+ waves indicate a half-time in the order of 15–30 minutes for Gap26 and 30–60 minutes for Gap27 (Boitano and Evans, 2000). By contrast, exposure times as long as 6 hours have been reported to have no effect on FRAP-based dye coupling (Decrock et al., 2009a) and 24 hours is necessary to see an effect on coupling (De Bock et al., 2011). At the other extreme are observations based on capacitance measurements that indicate GJ inhibition within 5 minutes (Matchkov et al., 2006). Possibly, capacitance measurements may be influenced by effects on HCs, because those channels are typically blocked on a time scale of a few minutes (Desplantez et al., 2012b; Wang et al., 2012a) (HC effects are discussed below). It is currently not known how Gap26/27 interact with the ELs; there are no reports on direct molecular interactions based on surface plasmon resonance or NMR studies. Work with atomic force microscopy whereby Gap26 was covalently linked to the scanning tip has demonstrated interaction with the Cx43 ELs based on the unbinding force experienced by retracting the scanning tip after interaction was established (Liu et al., 2006). As pointed out earlier (see section II.B.2), HC docking involves EL1-EL1 and EL2-EL2 interactions between opposed connexins, but EL1-EL2 and EL2-EL1 interactions are also possible. Recently, the Green group performed competition experiments with Peptide5, a connexin HC-inhibiting peptide based on the EL2 SRPTEK (“Gap27”) motif (Fig. 1), with various neighboring EL2-based sequences (Kim et al., 2017). They found that Peptide5 inhibition was counteracted by equimolar addition of an EL2 peptide with a sequence N-terminally adjacent to the Peptide5 sequence, indicating that Peptide5 interacts with EL2. Most interestingly, a peptide composed of the entire EL1 sequence acted synergistically with Peptide5, giving significantly increased inhibition. Thus, EL2-derived peptides appear to interact with EL2 and not with EL1; it remains to be determined whether EL1-based peptides interact with EL1. Additionally, combination of EL1- and EL2-based peptides may provide stronger inhibition.
A second possibility for Gap26/27 effects may relate to direct interactions of these peptides with existing GJs, resulting either in the separation of HCs from existing GJs or in effects on GJ gating. Separation of docked HCs is in principle possible, and continuation of this process would result in unzippering of the GJ followed by internalization and breakdown. Solan and Lampe (2014) have suggested that the switch between unzippering and uptake of the complete GJ as an annular junction (see section II.B.4) is determined by distinct Cx43 phosphorylation patterns (Solan and Lampe, 2014). Although it is conceivable that Gap peptides may have access to the intercellular space and HC docking region via the edge of the junctional plaque, there is currently very little evidence for the unzippering scenario given the tenacity of adhesion of docked HCs. The other possibility is that Gap26/27 interaction with GJs results in effects on GJ channel gating properties. Such effect has been described for an EL2 peptide called P180-195 composed of amino acids Ser-180–Gln-195 of Cx43 (Fig. 1), which appeared to decrease the dwelling in the subconductance (residual) state without pronounced alterations to the main conductance state (Kwak and Jongsma, 1999). However, peptide exposure times were very long (overnight) in this study. In the absence of any additional evidence, the possibility of direct actions of the Gap peptides on the gating or unzippering of GJs remains open for further investigations.
Thus far, most evidence points to interaction of Gap26/27 peptides with HCs not incorporated in GJs (see Fig. 3 in Leybaert et al., 2003). Most notably, Gap26/27 peptides inhibit HCs faster than GJs: macroscopic current measurements have suggested HC inhibition within 5 minutes (Desplantez et al., 2012b) and analysis at the unitary current level has demonstrated a half-time of ∼100 seconds (time constant τ of 148 seconds) and ∼150 seconds (τ = 223 seconds) for Gap26 and Gap27, respectively (Wang et al., 2012a). Removal of inhibition occurs within a time frame of 2.6 minutes (Desplantez et al., 2012b). As discussed earlier, GJ inhibition takes tens of minutes to hours, indicating that the peptides first interact with HCs and subsequently prevent the docking process, thereby inhibiting GJs. Although HC block occurs within minutes, this is still slow in terms of action dynamics on a channel protein. One reason could be that the interaction site on the ELs is not immediately accessible and needs some molecular rearrangements or unfolding. Another possibility is that there are other interactions with connexin domains that are buried deeper into the channel pore or that are only accessible when the channel is open; the latter option still needs to be tested by evaluating the use dependence of block. In fact, the time-dependence curve of HC block in Wang et al. (2012a) may contain some use-dependent effect, because the protocol of this experiment involved repetitive application of HC opening voltage steps. In terms of mechanisms of HC inhibition, the classic possibilities are decreased number of channels, pore block, decreased unitary conductance, or effects on gating (reviewed in Wang et al., 2013a). Effects on the number of channels and single channel conductance have been excluded (Wang et al., 2012a), at least for an exposure time of 30 minutes. Pore block occurs at high concentrations, in the order of 1 mM, in which case control peptides (e.g., with a scrambled sequence) as well as active peptide inhibit HCs (Wang et al., 2012a), demonstrating that block becomes unspecific at these high concentrations. The most probable effect of Gap26/27 resides at the level of channel gating: Wang et al. (2012a) reported that Gap26/27 decreases the dwelling in the main conductance state without affecting the subconductance state (Wang et al., 2013a); these findings are opposite of the effects reported by Kwak and Jongsma (1999) with the P180-195 EL2 peptide (Fig. 1), which demonstrated decreased dwelling in the subconductance state without changes for the main conductance state. As transitions from fully closed to the main open state involve the slow loop gate, it can be concluded that Gap26/27 inhibition results from closure of the loop gate. Gap26/27 also have other effects on gating, in that they increase the Vm threshold for voltage activation. Although substantial mechanistic insights have been obtained on mechanisms of Gap26/27 inhibition of electrically triggered HC opening, little experimental evidence is available on their effect on gating mechanisms involved in chemically triggered HC opening. However, as the slow loop gate plays a central role in chemical gating, it looks obvious to suppose that Gap26/27 would also inhibit loop gate opening triggered by, e.g., [Ca2+]i elevation or exposure to proinflammatory or ischemic conditions. Just like is the case for GJs, Gap26/27 block of HCs is incomplete. Gap26 HC block appears to be less complete (∼65% inhibition) than observed with Gap27 (∼84% inhibition; at supramaximal concentrations). On the other hand, the IC50 of Gap26 is lower (∼80 µM) than for Gap27 (∼160 µM) (Wang et al., 2012a). Note that the IC50 of these peptides for HC inhibition are higher than those reported by Chaytor et al. (1997) for inhibition of GJs; however, as mentioned earlier, the assay in the Chaytor et al. (1997) study was a very indirect one based on rhythmicity of blood vessel rings exposed to phenylephrine, whereas the HC assays of Wang et al. (2012a) were based on measurements of unitary HC opening activities. Although Gap26/27 peptides have some specificity for different connexin isoforms, their selectivity has been challenged, because they were found also to inhibit Panx1 channels (Dahl, 2007; Wang et al., 2007a). Limited effects of Gap26/27 peptides on other proteins like Panx1 channels are not unexpected, given the high concentration of 200 µM or higher needed to block connexin channels; however, the effect of Gap26/27 on connexin channels is much stronger than on Panx1 channels. It is therefore advised to use concentrations less than 200 µM, because nonspecific effects will progressively appear; at 1 mM, nonspecific block is maximal with no difference between active and scrambled peptide sequences (Wang et al., 2012a).
b. Peptide5.
Peptide5 is a peptide based on the SRPTEK domain with sequence VDCFLSRPTEKT (Fig. 1) first reported by O’Carroll et al. (2008). This peptide came out as most potent after a screen of several peptides mimicking sequences on the Cx43 ELs (including some containing TM portions) for their potential to inhibit cell swelling, astrogliosis, and neuronal cell death in an in vitro model of spinal cord injury. They found that Peptide5 inhibited HCs at concentrations of 5 µM, whereas hundreds of micromolars were necessary to inhibit GJs. Follow up work demonstrated significant neuroprotective and inflammation dampening effects of this peptide in various models including brain ischemia in fetal sheep (Davidson et al., 2012b, 2014), retinal ischemia (Danesh-Meyer et al., 2012), and spinal cord injury (O’Carroll et al., 2013) and subsequently extensively tested for its neuroprotective potential (see section VIII). Recent work from Kim et al. (2017) demonstrated that HC block is very sensitive to alterations of the Peptide5 sequence (either single amino acid substitutions for Ala or truncations at the NT or CT side) and that the SRPTEKT sequence was not sufficient on its own to block HCs (Kim et al., 2017). By contrast, GJ block was not significantly altered by these modifications and SRPTEKT acted equally well as Peptide5. This suggests distinct interactions sites for Peptide5 inhibition of HCs and GJs. Most interestingly, combining Peptide5 with EL1 peptide (entire loop sequence) increased the potency of HC inhibition. Given the fact that HC inhibition (but not GJ inhibition) is very sensitive to sequence alterations of Peptide5, including single amino acid substitutions outside the SRPTEKT sequence, suggests a possible sequence specificity for HC inhibition, an option that needs to be further tested. Peptide5 inhibition of GJs appeared to be mediated by altered distribution of Cx43 without influencing the expression level (Kim et al., 2017). Currently there are no data available on Peptide5 effects at single channel resolution to judge its effect on unitary conductance and gating.
2. Mimetic Peptides of Intracellular Connexin Sequences.
a. GAP19 and L2-specific hemichannel blocking peptides.
Although Peptide5 inhibits HCs only at low concentrations and gives combined GJ/HC inhibition at high concentrations, Gap19 and L2 peptides inhibit HCs while they prevent the closure of GJs, i.e., they have opposite effects on GJs and HCs (Fig. 2A). Gap19 is a peptide mimicking a nine amino acid sequence on the CL of Cx43 located within the L2 sequence (Fig. 1). Combined work of the Bultynck and Leybaert groups has demonstrated that L2 peptide, which was known to prevent GJ closure upon acidification (see section II.E.2) unexpectedly inhibited Cx43 HCs (Ponsaerts et al., 2010; reviewed in Iyyathurai et al., 2013). They further identified Gap19 peptide as a sequence within the L2 domain that is flanked at both sides by α-helices at pH 5.8 (Duffy et al., 2002). In previous work with Gap26/27, mimetic peptides of the CL like Gap19 were used as inactive control peptides for GJ studies. In line with this, Gap19 had no acute effect on Cx43 GJs while it slightly promoted GJ coupling when applied for 24 hours or more (Wang et al., 2013c). Similar findings have been reported for L2 peptide, which decreases the frequency of transitions from the main to the residual state of GJ channels (Seki et al., 2004). Surprisingly, it was found that Gap19, like L2 peptide, inhibited Cx43 HCs (Wang et al., 2013c). They further demonstrated by surface plasmon resonance experiments that Gap19 interacted with the CT and that CT9 peptide inhibited Gap19 HC inhibition in a dose-dependent manner, indicating that Gap19 prevents CL interaction with the CT9 region. As explained previously, CT-CL interaction is necessary for Cx43 HCs to become available for opening (see section II.E.5). As a result, Gap19 prevention of CT-CL interaction will bring Cx43 HCs into a state where they are unavailable for opening. Interaction of Gap19 with the CT target is characterized by a Kd of ∼2.5 µM. Although the CT9 domain is a target of Gap19, other CT sites may also be involved, as was observed with L2 peptide in the context of GJ coupling (see section II.E.2). Gap19 has some intrinsic membrane permeability but coupling of Gap19 to the TAT translocation sequence strongly increases its membrane permeability; experiments with TAT-Gap19 indicated an IC50 of ∼7 µM, i.e., close to the Kd for its interaction with the CT. Based on loss of channel function with the Cx43 I130T mutant (Shibayama et al., 2005) (Fig. 1), a mutation associated with ODDD (see section VII), an I130A mutant Gap19 version was tested and found to be inactive and thus useful as a control peptide (Wang et al., 2013c). Gap19 was tested on Cx40 HCs and Panx1 channels, on which it had no effects (Wang et al., 2013c). Of note, Gap19 contains several charged residues (4 Lys and 1 Glu) on a total of nine amino acids and its interaction with the CT may thus involve substantial electrostatic interactions. The interaction spectrum of the much longer L2 peptide (26 amino acids) will certainly be more elaborate and may ascertain specificity. L2 does not enter the cells spontaneously and needs to be coupled to a translocation motif such as TAT; mutant H126K/I130N L2 is the corresponding inactive control peptide (Ponsaerts et al., 2010). Interestingly, Gap19 has been used to investigate the role of Cx43 HCs in isoproterenol-induced cardiac arrhythmias, demonstrating a drastic improvement of isoproterenol-challenged animal mortality (González et al., 2015). Work with this peptide in mice has furthermore demonstrated it modestly reduces infarct size after myocardial ischemia/reperfusion (Wang et al., 2013c). TAT-linked Gap19 was shown to impair spatial short-term memory when injected into mice brain ventricles (Walrave et al., 2016). TAT-linked L2 peptide potently inhibited norepinephrine-induced vasoconstriction of rat small mesenteric arteries (Bol et al., 2016) and was demonstrated to inhibit fear memory consolidation when injected into rat basolateral amygdala (Stehberg et al., 2012).
Just like [Ca2+]i, proinflammatory cytokines and metabolic inhibition distinctly influence HCs and GJs, this also appears to be the case for Gap19 and L2 peptides. Taking into account that both peptides interfere with CT-CL interaction and are identical to the sequence of domains crucial for this interaction, this suggests that the process of docking and GJ formation induces conformational changes that switch the outcome of CT-CL interaction from necessary for HC opening to inhibition of GJ.
Interestingly, the L2 peptide has been further explored in terms of its pharmacophore for interacting with the Cx43 CT, with the RXP-E motif appearing as an interesting starting point from which other peptides and petidomimetics have been developed (see section II.E.2) (Shibayama et al., 2006b; Lewandowski et al., 2008; Verma et al., 2009). In particular, the linear peptide RRNYRRNY, the cyclic peptide CyRP-71, and the peptidomimetic molecule ZP2519 were demonstrated to target the Cx43 CT and to prevent Cx43-based GJ closure under low pH conditions (Verma et al., 2009,2010). These substances are of potential translational value for preventing postischemic GJ closure. Moreover, these molecules are potential HC blockers and may thus have two-sided actions directed at preventing GJ closure as well as inhibiting HC opening. Interestingly, RRNYRRNY was recently demonstrated to act as a Cx43 HC blocker, not only of plasma membrane HCs but also of HCs in subsarcolemmal mitochondria (Gadicherla et al., 2017). As a result, RRNYRRNY displays three levels of action on Cx43-based channels: prevention of GJ closure, inhibition of HCs in the plasma membrane, and inhibition of mitochondrial HCs (mitochondrial connexin channels are further discussed in section V.A).
2. JM2 Peptide.
Recently, the Gourdie group reported that JM2 peptide, composed of 15 amino acids in the juxtamembrane region of the Cx43 CT (see Fig. 1), inhibited ATP release triggered by low extracellular Ca2+ in human microvascular endothelial cells (Calder et al., 2015). The peptide overlaps with the crucial juxtamembrane (JM) domain (amino acids 234–243, indicated in Fig. 1) that functions as a microtubule binding site, which is part of a larger domain stretching from amino acid 228 to 263 (Giepmans et al., 2001). JM2 peptide was tested on hemichannel ATP release, but its effects on gap junctions were not investigated.
3. CT9 Peptide
As discussed earlier, CT9 peptide removes the high [Ca2+]i brake on Cx43 HC opening (see section II.E.5). As a result, CT9 acts as a promoter of [Ca2+]i-linked HC opening (Fig. 2B), as illustrated recently in vascular smooth muscle cells (Bol et al., 2016). CT10, which contains an additional Ser (Ser-373), has the same effect (Ponsaerts et al., 2010), but CT9 is the smallest active sequence (the Ser was added as a linker to the TAT translocation motif). Moreover, CT9ΔI peptide that lacks the last CT-located Ile (Ile-382 in the Cx43 protein) and therefore does not compete with Cx43–ZO-1 interaction (see section II.B.3) still displays its HC activity-enhancing effect (De Bock et al., 2012). Similar to Gap19, L2, and RRNYRRNY, CT9 has its target inside the cell (L2/Gap19 sequences on the CL) and needs to be fused to a translocation sequence to improve its membrane permeation (TAT-CT9).
αCT1 is another form of CT9 peptide, which is linked to the antennapedia membrane translocation sequence instead of the TAT sequence. Interestingly, αCT1 promotes HC incorporation into GJs and increases plaque size by facilitating HC disconnection from ZO-1–linked cytoskeletal elements as a result of competition between the peptide and the endogenous Cx43 CT9 motif (Hunter et al., 2005; Palatinus et al., 2012; Rhett et al., 2013) (see section II.B.3). αCT1 reduces inducible arrhythmias after cryoinjury applied to the left ventricle in the heart and prevents Cx43 lateralization/remodeling; it also increases Cx43 phosphorylation at Ser-368 in a PKCε-dependent manner (O’Quinn et al., 2011). These effects are exerted not within seconds per minute characteristic for CT9 effects on HC gating but on a time scale of hours that is characteristic for the cardiac Cx43 life cycle. As a result CT9, fused to either TAT or the antennapedia sequence, has two effects: within minutes it promotes [Ca2+]i-linked HC opening, whereas within hours it promotes HC incorporation into GJs, thereby reducing the HC pool in the plasma membrane. If CT9 is used with the purpose of promoting HC incorporation into GJs (long-term effect to improve GJ coupling), it is advisable to consider combining its administration together with an HC blocker to prevent the potentially inappropriate acute HC-opening effect of CT9-based peptides.
4. Note on Hemichannel-Targeting Strategies.
Currently Gap19, L2, and RRNYRRNY peptide (Fig. 1) are the only known molecules that inhibit Cx43 HCs without inhibiting GJs. Peptide5 is an HC blocker at low (5 µM) concentrations that blocks GJs at higher concentrations; Gap26/27 inhibit HCs within minutes, with delayed effects on GJs (hours). Two other (nonpeptide) compounds namely boldine (Hernandez-Salinas et al., 2013) and D4 (Cea et al., 2016b) have been reported to act as connexin HC blockers. Boldine is an alkaloid extracted from the boldo tree but the chemical identity of D4 was not revealed. Recently, Yi et al. (2017) demonstrated that boldine inhibited HCs but not GJs in astrocytes at concentrations of 0.1 and 0.5 mM in cultures and acute brain slices. These authors furthermore showed that long-term in vivo boldine administration (12 weeks, administered in the drinking water) inhibited HCs but not GJs; in APP/PS1 mice such treatment reduced [Ca2+]i elevation in astrocytes, tempered gliotransmitter release and mitochondrial superoxide ion load, and reduced dystrophic neurite counts (Yi et al., 2017). Amyloid load was not affected and behavior/cognition were not investigated.
Certain glycyrrhetinic acid-based molecules have been claimed to act in an HC-specific manner but there is little convincing evidence to support this claimed selectivity (Takeuchi et al., 2011). Lanthanum ions (La3+) block HCs and indeed do not inhibit GJs (Anselmi et al., 2008); however, this ion has manifold other effects, including the inhibition of Ca2+ channels (Mlinar and Enyeart, 1993; Young et al., 2002).
Besides peptides, two ODDD Cx43 mutants have been characterized to display increased HC function combined with decreased gap junctional function. These mutants are G138R and G60S (see Fig. 1) (Dobrowolski et al., 2007, 2008; Kozoriz et al., 2013). As such, these mutants are interesting tools to distinguish HC function from GJ function. The G138R mutant has been used to investigate the consequences of increased HC function in brain slice experiments (Torres et al., 2012), whereas the G60S mutant was demonstrated to increase infarct size in animal stroke models (Kozoriz et al., 2013). Another interesting mutant not linked to ODDD is R76W, which has impaired GJ function but preserved HC function (Xu et al., 2015) and G8V that has preserved GJ function but increased HC opening (Wang et al., 2015a) (Fig. 1).
3. Peptides not Mimicking Connexin Sequences: The Case of AAP10.
The most striking and best characterized peptide with pronounced effects on GJ channels, which is not based on a connexin protein sequence, is AAP10 peptide and analogs. AAP10 stands for antiarrhythmic peptide 10 and is derived from a naturally occurring antiarrhythmic peptide isolated from bovine atria, which enhanced the synchronization of cultivated clusters of spontaneously beating embryonic chicken cardiomyocytes (Aonuma et al., 1980). Starting from this original antiarrhythmic peptide, a number of synthetic derivatives was developed, among which AAP10 (GAG-4Hyp-PY) (illustrated in Fig. 3) was most active and chosen as lead substance (Grover and Dhein, 2001). The hypothesis that “enhanced synchronization of cardiomyocyte clusters” may represent an increase in intercellular coupling could be verified: simulated ischemia (hypoxia combined with glucose deficit) leads to a reduction in conduction velocity in guinea pig papillary muscles that could be prevented by 1 µM AAP10 added to the bath solution (Müller et al., 1997a). Furthermore, it was found that in dual whole cell patch clamp experiments, concentrations of 10 or 50 nM AAP10 lead to enhanced electrical coupling between pairs of cardiomyocytes without affecting the sodium, calcium, potassium, or other transmembrane ionic currents (Müller et al., 1997a,b; Weng et al., 2002; Hagen et al., 2009). Similarly, enhanced coupling was also observed with rotigaptide, a peptide that is further discussed below (Jørgensen et al., 2005; Clarke et al., 2006). Further experiments showed that the positive effect of AAP10 on electrical coupling was increased if cells were partially uncoupled: cells exposed to low pH (6.5 via CO2) exhibit reduced GJ conductance that was prevented by pretreatment with 50 nM AAP10 and, moreover, could also be reversed by treatment with AAP10 (Hagen et al., 2009). Besides electrical coupling, AAP10 also increased metabolic GJ coupling as evident from increased dye transfer (Hagen et al., 2009). The effects of AAP10 on GJs could be demonstrated in rat (Hagen et al., 2009), guinea pig (Müller et al., 1997a,b; Weng et al., 2002), rabbit (Dhein et al., 1994; Jozwiak and Dhein, 2008), and human (Hagen et al., 2009) cardiomyocytes. Experiments using HeLa cells stably transfected with either Cx43, Cx40, or Cx45, i.e., the typical cardiac connexins, showed that AAP10 acts on GJs via effects on Cx43 or Cx45 but not via Cx40 (Easton et al., 2009; Hagen et al., 2009).
Chemical structure of AAP10.
Because peptides often are biologically unstable, further research was focused on the structural chemistry of these peptides and on development of a radioligand binding assay to allow the development of nonpeptide drugs. Structure-activity relations together with molecular modeling, and two-dimensional NMR spectroscopy revealed that AAP10 has a semicyclic structure like a horseshoe, which is due to the two Pro residues (Dhein et al., 2010; Grover and Dhein, 1998, 2001). Moreover, it could be shown that [14C]AAP10 binds to a membrane protein of rabbit cardiomyocyte membranes (Dhein et al., 2001) with a Kd in the order of 0.3–0.9 nM and maximum binding Bmax in the order of ∼42.5 pmol/mg (Jozwiak et al., 2012). The finding of the semicyclic structure and the identification of the essential chemical moieties (Grover and Dhein, 1998, 2001) led to the idea of replacing some of the amino acids by d-amino acid versions. To maintain certain groups at their positions, the order of these amino acids had to be reversed, which, using a retro-all-d amino acid design of the AAP10 template, led to ZP123 (YP-4Hyp-GAG) also named rotigaptide (Kjølbye et al., 2003; Xing et al., 2003). Due to the d-amino acids, its in vivo half-life is longer than that of AAP10. As AAP10, the d-amino acid analog rotigaptide enhanced GJ communication but did not bind to other transmembrane ion channels (Haugan et al., 2005). Further reduction to the pharmacophore allowed synthesis of the dipeptide danegaptide also called ZP1609 (Skyschally et al., 2013) and of the nonpeptide drug Gap134 [(2S,4R)-1-(2-aminoacetyl)-4-benzamido-pyrrolidine-2-carboxylic acid hydrochloride] (Butera et al., 2009).
Until this point, we have only considered the effects of these drugs on GJs, but the question arises whether antiarrhythmic peptides really exert antiarrhythmic effects and against which type of arrhythmia they may be effective. Early considerations using computer simulations about the relationship of intercellular coupling and arrhythmia started with the idea that reduced coupling might unmask local inhomogeneities in action potential duration (APD), resulting in a dispersion of APDs (Lesh et al., 1989; Müller and Dhein, 1993). Together with slowed conduction these local differences in refractoriness could lead to reentrant arrhythmia (for more detail see section VI). Based on this hypothesis, 256 electrode mapping was performed using isolated rabbit hearts, which demonstrated, in good accordance to the computer simulations, that under control conditions AAP10 significantly reduced dispersion in APD (Dhein et al., 1994) and that regional ischemia results in alterations of the spread of activation that were reduced by AAP10 (Dhein et al., 1994). In addition, ischemia-induced slowing of conduction in the ischemic border zone was antagonized by AAP10 (Jozwiak and Dhein, 2008). Accordingly, it was also shown that AAP10 could significantly decrease the incidence of ventricular tachycardia and ventricular fibrillation in acute ischemia (coronary ligation) (Ni et al., 2015; Sun et al., 2015a). Similarly, rotigaptide (ZP123) also reduced the incidence of ventricular fibrillation and ventricular tachycardia in acute coronary ischemia in various animal models (Hennan et al., 2006; Kjølbye et al., 2008; Su et al., 2015), prevented ischemia-induced conduction velocity slowing (Shiroshita-Takeshita et al., 2007) and suppressed arrhythmia in volume-pressure overload heart failure (Liu et al., 2014). Spiral wave reentry circuits became destabilized by enhancement of GJ coupling (Takemoto et al., 2012). As AAP10, rotigaptide also reduced APD dispersion (Dhein et al., 2003). Interestingly, AAP10 additionally reduced the inducibility of ventricular fibrillation in healed myocardial infarction (Ren et al., 2006). Of note, rotigaptide also reduced infarct size (Haugan et al., 2006; Hennan et al., 2006), although others did not observe this effect (Xing et al., 2003). Infarct size reduction has also been found with danegaptide (ZP1609) (Skyschally et al., 2013). Recent work demonstrated ZP1609 prevents cardiomyocyte hypercontracture after ischemia-reperfusion by acting on mitochondria (Boengler et al., 2017); this particular effect was, however, not linked to Cx43. AAP10 has furthermore been demonstrated to prevent drug-induced torsade de pointes arrhythmia and early afterdepolarizations (Quan et al., 2007, 2009; Ruan et al., 2014). In cardiac preparations uncoupled with lysophosphatidic acid, AAP10 antagonized ventricular tachycardia induced by programmed S1S2 stimulation (Zhou et al., 2011). In vivo, the inducibility of ventricular fibrillation by the plant toxin aconitine was significantly attenuated by AAP10 (Dhein et al., 2001).
In contrast, AAP10 did not affect burst stimulation-induced atrial fibrillation (Haugan et al., 2004). Similarly, rotigaptide did not prevent atrial tachyarrhythmia development in a chronic volume overload rabbit model (Haugan et al., 2006), whereas in a canine mitral regurgitation model, rotigaptide was effective in reducing atrial fibrillation inducibility (Guerra et al., 2006) as well as against atrial conduction slowing induced by metabolic stress (Haugan et al., 2006) or atrial stretch (Ueda et al., 2014). This was also found for Gap134 (Rossman et al., 2009). AAP10 could not reverse diabetes mellitus-induced conduction slowing in Zucker diabetic fatty rats (Olsen et al., 2013); this may be linked to the fact that this type of conduction slowing is probably due to enhanced fibrosis rather than resulting from functional electrophysiological changes and may therefore be more difficult to reverse. Taken together, most researchers found evidence for antiarrhythmic effects of AAP10 and related peptides against ischemia-induced ventricular tachyarrhythmias, while efficacy in atrial fibrillation seems to depend on the model used and the type of atrial fibrillation. The efficacy of antiarrhythmic peptides seems to be highest in partially uncoupled tissue if uncoupling is induced by hypoxia, ischemia, or acidosis.
This leads to the question of the biochemical mechanisms of action of antiarrhythmic peptides. First of all, it has been shown that AAP10 and rotigaptide binds to a membrane protein with a Kd in the range of 0.1–0.9 nM [[14C]AAP10 (Dhein et al., 2001; Jozwiak et al., 2012); [125I]di-I-AAP10 (Jørgensen et al., 2005)]. A 200-kDa membrane protein could be isolated from cardiac tissue by affinity chromatography and cross-linking techniques (Weng et al., 2002), but its identity was not determined. Regarding the subsequent signal transduction cascade, the antagonization of the AAP10 effects by GDP-βS is in favor of the idea that G-proteins are involved in the signaling process (Weng et al., 2002). Furthermore, the peptides activate protein kinase Cα [AAP10 (Dhein et al., 2001; Weng et al., 2002); rotigaptide (Dhein et al., 2003)]. The effect of AAP10 on GJIC and on Cx43 phosphorylation could be completely inhibited by blocking PKC (Weng et al., 2002; Easton et al., 2009), indicating that PKC is critically involved in the signal transduction process. Consistently, several groups found enhanced phosphorylation of Cx43 after incubation of the cells/tissue with antiarrhythmic peptides [AAP10 (Dhein et al., 2001; Weng et al., 2002; Quan et al., 2007, 2009; Wang et al., 2007b; Easton et al., 2009; Sun et al., 2015a); rotigaptide (Kjølbye et al., 2008; Su et al., 2015)]. In contrast, others did not observe enhanced Cx43 phosphorylation with rotigaptide (Clarke et al., 2006). However, within the first 30 minutes of ischemia, Ser-306, Ser-297, and Ser-368 of Cx43 are dephosphorylated, whereas Ser-330 is phosphorylated. The dephosphorylation of Ser-297 and Ser-368 can be prevented by rotigaptide (Axelsen et al., 2006). Cx43 synthesis was not affected by rotigaptide (Liu et al., 2014). In contrast, AAP10 led to an increase in Cx43 mRNA expression (Easton et al., 2009).
More detailed analysis revealed that the effects of AAPs on Cx43 phosphorylation were enhanced in ischemic tissue and attained a maximum in the ischemic center while less pronounced in nonischemic tissue (Jozwiak and Dhein, 2008), which may explain the controversial findings cited above. Accordingly, others found that the binding site density for AAP10 was enhanced during metabolic stress (Jørgensen et al., 2005). Thus, one might conclude from these finding that AAPs prevent from dephosphorylation rather than induce active phosphorylation.
Phosphorylation of Cx43 CT can alter the single channel conductance (Takens-Kwak and Jongsma, 1992; Moreno et al., 1994; Kwak et al., 1995a; Kwak and Jongsma, 1996), but also is critically involved in controlling the transfer and insertion of Cx43 into the membrane (involving the tubulin apparatus) and in the removal of Cx43 from the membrane (Lampe and Lau, 2000; Solan and Lampe, 2007; Saidi Brikci-Nigassa et al., 2012). Accordingly, two groups found that Cx43 density in the membranes is reduced in ischemia and can be preserved by AAPs (Jozwiak and Dhein, 2008; Sun et al., 2015a). Moreover, it was shown by these authors that the reduction of Cx43 and its inhibition by AAP10 was highest at the cell poles. Interestingly, AAP10 led to a higher Cx43 density at points of cell-cell contact (Easton et al., 2009), which would be in line with the assumption that the incorporation into the membrane might be enhanced. The mechanism of action of the AAPs is summarized in Fig. 4.
Synthesis, transport, and membrane incorporation of cardiac connexins and mechanisms of ischemic damage leading to GJ closure, connexin dephosphorylation, and connexin-removal from the membrane. Molecular mechanisms and effects of antiarrhythmic peptides on GJ channels formed by Cx43 or Cx45 are also illustrated. GDP-βS, guanosine-5′-[β-thio]diphosphate trilithium salt (nonhydrolyzable GDP analog); PKCα, protein kinase Cα; CGP 54345, PKC inhibitor (inhibits only PKCα); HBDDE 2,2′,3,3′,4,4-hexahydroxy-1′-biphenyl-1-6,6′-dimethanoldimethylether, inhibitor of PKCα and PKCγ; BIM I, bisindolyl-maleimide I, inhibitor of PKCα; βI, βII, γ, δ, and ε; SR, sarcoplasmic reticulum; AAPnat, natural antiarrhythmic peptide (H-Gly-Pro-Hyp-Gly-Ala-Gly). Modified from Dhein et al. (2010).
Regarding clinical safety, rotigaptide was evaluated in a phase I study in 200 healthy subjects and found to be safe (Kjølbye et al., 2007). Taken together, AAPs have been demonstrated to possess a pharmacological potential for the treatment of ischemia- or hypoxia-linked arrhythmias related to cellular uncoupling via Cx43 or Cx45. On the background of the observations that GJs also control growth and differentiation of cells, this approach may also be worth investigating in the area of cancer (Salameh and Dhein, 2005).
IV. Connexins in Vascular Disease
Atherosclerosis, a progressive inflammatory disease of large and medium-sized arteries, is the number one killer worldwide. The main complications of atherosclerosis, namely ischemic heart disease and stroke, are the world’s first and third leading causes of death, representing 28.5% of all-cause mortality (GBD 2013 Mortality and Causes of Death, 2015). The disease involves the formation of plaques in the intima of arteries that are characterized by a dysfunctional endothelium, leukocyte and lipid accumulation, cell death, and fibrosis (Fig. 5). Atherosclerotic plaques develop predominantly at arterial locations where regular (high) laminar blood flow is disturbed, i.e., arterial bifurcations and branch points (Kwak et al., 2014). The most severe clinical events follow the rupture of a plaque (Fig. 6), which exposes prothrombotic material inside the plaque to the blood and causes sudden thrombotic occlusion of the artery at the site of disruption (Hansson et al., 2015). Various connexins have been shown to be involved in the initiation and progression of atherosclerosis. Furthermore, these proteins may also influence thrombus formation and stabilization.
Connexin expression in healthy arteries (A), stable atherosclerotic plaques with a thick fibrous cap (B), and after rupture of vulnerable lesions. Connexin expression is represented according to cell type. DCs, dendritic cells; ECs, endothelial cells; FC, fibrous cap; LC, lipid core MCs, monocytes; MFCs, macrophage foam cells; Plts, platelets; SMCs, smooth muscle cells; TC, T cells.
Connexin expression after myocardial ischemia (due to coronary occlusion) and reperfusion in wild-type mice. (Left) Cx37, Cx40, and Cx43 expression (in green) in unaffected healthy (H) myocardium. (Right) Cx37, Cx40, and Cx43 expression (in green) at the border zone (B) of the infarcted area (I). Tissue is counterstained with Evans Blue (in red), and nuclei are stained with DAPI (in blue).
Treatment of acute coronary atherothrombosis consists of procedures that allow the rapid return of blood flow to the ischemic zone of the myocardium to rescue heart muscle. However, reperfusion may paradoxically lead to further complications involving acceleration of cardiac cell death, diminished contractile function, and arrhythmias. The involvement of cardiomyocyte connexins in reperfusion injury will be described in section IV.C; here we will summarize recent evidence indicating an involvement of connexins in the inappropriate inflammatory response in the microcirculation that may be at the basis of ischemia-reperfusion injury (Bulluck et al., 2016).
Restenosis is the pathophysiological process involving inflammatory activation and intimal smooth muscle cell accumulation that occurs in about 10% of patients submitted to revascularization procedures of coronary, carotid, and peripheral arteries. It is in fact an excessive healing reaction of the vascular wall subjected to balloon angioplasty and endovascular stent implantation (Chaabane et al., 2013). Drug-eluting stents and the more recent drug-eluting balloons have significantly reduced but not eliminated the incidence of restenosis. The role of connexins in this clinically still relevant problem will be discussed.
A. Connexins in Atherosclerosis
There is growing evidence that connexins participate in the development of atherosclerotic disease. Early support for this hypothesis came from studies analyzing atherosclerotic lesions at different disease stages in specimens of human, rabbit, or mouse origin. First, Cx43 is generally absent in the endothelium of large arteries, but it has been found in endothelial cells at the shoulder region of advanced atherosclerotic plaques (Fig. 5) (Kwak et al., 2002). Although endothelial Cx43 expression may be induced by cytokines like TGF-β that are present in excess in an atherosclerotic environment (Larson et al., 2001), Cx43 expression may already have been upregulated at this specific arterial location before the atherosclerotic lesion formed. Support for the latter premise comes from the fact that abundant Cx43 expression has been observed in rat aortic endothelial cells localized at the downstream edge of the ostia of branching vessels and at flow dividers, regions known to experience disturbed blood flow (Gabriels and Paul, 1998). Subsequently, a causal relation between the induction of endothelial Cx43 and disturbed (or oscillatory) flow has been established in various in vitro studies (DePaola et al., 1999; Kwak et al., 2005; Feaver et al., 2008). Secondly, a high level of Cx43 expression has been found in intimal macrophages and smooth muscle cells of young atheroma, whereas Cx43 expression is downregulated in intimal smooth muscle cells of more advanced lesions (Fig. 5) (Polacek et al., 1993, 1997; Blackburn et al., 1995; Kwak et al., 2002). Interestingly, the oxidized phospholipid derivative POVPC reduces Cx43 levels of vascular smooth muscle cells, enhances its phosphorylation at Ser-279/282, and increases smooth muscle cell proliferation both in vitro and in an atherosclerotic mouse model in vivo (Johnstone et al., 2009). Finally, Cx37 and Cx40 levels are also modified during the course of atherosclerotic disease in human and mice (Kwak et al., 2002). Thus, Cx40 and Cx37 disappear from endothelial cells covering advanced atherosclerotic lesions, and Cx37 expression is enhanced in macrophage foam cells (Fig. 5). Moreover, long-term hyperlipidemia (a well-known atherogenic condition) reduces endothelial Cx37 and Cx40 expression in mouse aorta, an effect that could be reversed only for Cx37 by 1 week of treatment with simvastatin, a cholesterol-lowering drug (Yeh et al., 2003). In addition, endothelial Cx37 is downregulated in response to disturbed flow (Pfenniger et al., 2012b). Altogether, these studies brought the idea that connexin expression or posttranslational modifications in connexins might evolve in atherosclerotic plaques over time, depending on the stage of the lesion and might thus affect atherogenesis.
As germ-line loss of Cx43 is lethal (Reaume et al., 1995), the role of Cx43 in atherosclerosis was first studied in Cx43+/− mice crossed with atherosclerosis-susceptible low-density lipoprotein (LDL) receptor-deficient (Ldlr−/−) mice fed a high-cholesterol diet. Initial studies on Cx43+/−Ldlr−/− mice revealed that Cx43 has an overall atherogenic effect, and that reducing Cx43 might be beneficial by both reducing plaque burden as well as stabilizing the lesions (Kwak et al., 2003). However, the exact scenario by which ubiquitous reduction in Cx43 leads to this dual benefit remained uncertain due to Cx43 expression in multiple atheroma-associated cell types (Wong et al., 2003). To examine specifically the role of Cx43 in immune cells, Ldlr−/− mice were lethally irradiated and reconstituted with Cx43+/+, Cx43+/−, or Cx43−/− hematopoietic fetal liver cells (Morel et al., 2014b). Intriguingly, the progression of atherosclerosis was lower in Cx43+/− chimeras compared with Cx43+/+ and Cx43−/− chimeras, and their plaques contained fewer neutrophils. It turned out that chemoattraction of neutrophils, which themselves do not express Cx43, was reduced in response to supernatant secreted by Cx43+/− macrophages in comparison with the ones of Cx43+/+ and Cx43−/− macrophages. Thus, titration of Cx43 levels in macrophages might regulate their chemoattractant secretion, leading to a reduction in atherosclerosis (Morel et al., 2014b).
In contrast to Cx43, Cx40 protects against atherosclerosis in mice by synchronizing endothelial CD73-dependent anti-inflammatory signaling, thus inhibiting leukocyte recruitment to the atherosclerotic lesion (Chadjichristos et al., 2010). Furthermore, loss of Cx37 promotes the development of atherosclerosis in apolipoprotein E-deficient (Apoe−/−) mice. Mechanistically, it appeared that ATP release through Cx37 HCs in monocytes control the initiation of atherosclerotic plaque development by regulating their adhesion (Wong et al., 2006). Recent in vitro studies show a downregulation of Cx37 expression, inhibition of ATP release, and augmented adhesion of the human monocytic cell line THP-1 by oxidized LDL (Liu et al., 2016). These harmful effects of oxidized LDL could be in part prevented by rutaecarpine, an active component of a Chinese herbal medication (Liu et al., 2016). Finally, Cx37 deletion in apolipoprotein E−/− (Apoe−/−) mice not only controls the initiation of atherosclerotic plaque formation but also increases the size of advanced lesions and abrogates the development of a stable plaque phenotype in regions exposed to oscillatory shear stress (Pfenniger et al., 2015), suggesting that local hemodynamic factors may modify the risk for Cx37-related adverse disease outcomes. A recent study examining the effects of lentiviral Cx37 interference on established abdominal aortic plaques in pigs by intravascular ultrasound revealed a reduction in plaque volume in the 8 months after lentiviral transduction, illustrating that not only local hemodynamic factors but also the disease stage itself may influence the outcome of reducing Cx37 on atherosclerosis burden (Guo et al., 2015). Although these animal studies have revealed important and diverse contribution of vascular connexins to atherogenesis, more work is needed to uncover the roles of these proteins in human disease.
During the past 15 years, a single nucleotide polymorphism (SNP) in the human Cx37 gene (GJA4) has been associated in a variety of populations with increased risk for various clinical manifestations of atherosclerosis, such as coronary artery disease, myocardial infarction, and ischemic stroke; however, these studies remained controversial as to which allele carried the risk (see for a review Meens et al., 2012). Consequently, data of multiple studies were extracted by two independent reviewers and a total of 3498 myocardial infarction cases and 3986 controls, as well as 1808 coronary artery disease cases and 1197 controls have been enrolled in a recent meta-analysis (Wen et al., 2014). This meta-analysis demonstrated that the GJA4-1019T allele is a risk factor for myocardial infarction and a protective factor for coronary artery disease. The GJA4 1019C>T SNP results in a nonconservative Pro to Ser substitution in the CT of Cx37, which has a significant impact on channel function under basal and phosphorylating conditions (Derouette et al., 2009; Morel et al., 2010; Pfenniger et al., 2010). In accordance with the above described studies using Cx37-deficient monocytes, monocytic cells expressing Cx37-319P, encoded by GJA4-1019C, were markedly less adhesive than cells expressing Cx37-319S, encoded by GJA4-1019T (Wong et al., 2006), thus suggesting that Cx37-319P polymorphic HCs may function as a protective genetic variant for plaque rupture, leading to myocardial infarction by specifically retarding recruitment of monocytes to human atherosclerotic lesions. In contrast to GJA4 1019C>T, SNPs located in the promoter region (−1930 C>T) or 3′-untranslated region (1297 I>D) of GJA4, which are presumed to affect Cx37 transcription level or mRNA stability, were not associated with altered risk for coronary artery disease (Han et al., 2008). Moreover, two SNPs in the promoter region of the GJA5 gene, −44G>A and +71A>G, that were found to significantly reduce Cx40 transcription (Firouzi et al., 2006) could also not be associated with an altered risk for coronary artery disease or acute myocardial infarction (Pfenniger et al., 2012a; Seifi et al., 2013). Although the latter studies remain to be confirmed in large cohorts or by meta-analyses, they stir up the idea that SNPs affecting connexin channel function may be of greater importance for cardiovascular disease than SNPs affecting connexin expression levels.
B. Connexins in Thrombosis
Cx37 was the first connexin found in platelets in 2011 (Angelillo-Scherrer et al., 2011), and the additional expression of a number of other connexins [Cx43, Cx40, and Cx32 (Vaiyapuri et al., 2012)] and Panx1 (Taylor et al., 2014; Molica et al., 2015) have been reported since. Transmission electron microscopy has convincingly revealed GJ-like structures between platelets (Vaiyapuri et al., 2012)). Moreover, platelets were shown to display functional Cx37 GJ channels during the aggregation response (Angelillo-Scherrer et al., 2011). Deletion of Cx37 in mice reduced tail bleeding time, shortened the time to occlusive arterial thrombosis, accelerated mortality in a model of thromboembolism, and enhanced platelet aggregation in response to modest concentrations of the agonists ADP, thrombin, and collagen (Angelillo-Scherrer et al., 2011). Furthermore, promotion of platelet aggregation in vitro was also observed with Cx37 mimetic blocking peptides. Given the biophysical properties of Cx37 GJ channels, it was hypothesized that these channels synchronize responses in platelets brought in contact during activation by transmitting cAMP to neighboring platelets, thereby functionally inhibiting freshly recruited platelets and limiting further thrombus growth (Angelillo-Scherrer et al., 2011). Interestingly, other studies report a loss of function in Cx37−/− as well as in Cx40−/− platelets, i.e., fibrinogen binding and α-granule secretion were decreased, even under conditions when direct platelet-platelet contacts were excluded (Vaiyapuri et al., 2012, 2013). This implies that the contribution of connexins to platelet aggregation may not be limited to the phase during which platelets come into stable contact with each other but might also occur when only HCs are present. The mechanism by which Cx37/Cx40 HCs open in the presence of relative high extracellular Ca2+ in blood (which normally keep HCs closed) still remains to be determined. Connexin HCs may allow for the release of ATP from activated platelets, which then might act in an autocrine/paracrine manner on P2X1, inducing a further increase in platelet activation state. The recent discovery of Panx1 in human platelets is in this respect of particular interest (Taylor et al., 2014; Molica et al., 2015).
C. Endothelial Connexins in Ischemia-Reperfusion Injury
It is increasingly recognized that deleterious consequences of ischemia-reperfusion are influenced by the dysfunction of the endothelium and the extent of neutrophil infiltration. Although the crucial role of endothelial connexins in diseases of large arteries is now well established, surprising little attention has been given to the role of endothelial connexins, i.e., Cx40 and Cx37, in the response of the microcirculation to ischemia-reperfusion injury (Fig. 6). Interestingly, spontaneous recovery of tissue perfusion after severe unilateral hindlimb ischemia was reduced, and the survival of distal limb tissue was compromised in Cx40−/− mice (Fang et al., 2012). The poor recovery from the ischemic insult in Cx40−/− mice appeared to be due to compromised regulation of tissue perfusion, vascular remodeling, and a prolonged inflammatory response (Fang et al., 2013). Thirty minutes of ischemia followed by 24-hour reperfusion resulted in increased myocardial infarct size in mice with endothelial-specific deletion of Cx40; however, no deleterious effects were found in mice with Cx37 deletion (Morel et al., 2014a). Mechanistically, endothelial Cx40-dependent cardioprotection appeared to involve CD73 activation that, in turn, limited neutrophil infiltration after cardiac reperfusion (Morel et al., 2014a). The Cx40-CD73 axis may thus represent a novel pharmacological target for controlling the damage associated with reperfusion in coronary disease.
The antiarrhythmic peptide rotigaptide or its analog danegaptide (see section III.B.3) reduce myocardial ischemia-reperfusion injury in procedures on large animals such as open-chest dogs (Hennan et al., 2006) and when given at the moment of reperfusion in pigs (Skyschally et al., 2013; Pedersen et al., 2016). Although the cardioprotective effects of these peptides are typically ascribed to the maintenance of Cx43-mediated GJ intercellular communication between cardiomyocytes, a recent translational study suggested additional effects of these peptides on the human endothelium. By measuring forearm arterial blood flow in humans, endothelium-dependent vasodilation in response to acetylcholine was attenuated after ischemia-reperfusion in the presence of placebo but not in the presence of intra-arterial rotigaptide (Pedersen et al., 2016).
D. Connexins in Restenosis
Acute vascular injury induced by balloon angioplasty is associated with increased Cx43 expression in neointimal smooth muscle cells and macrophages in various animal models (Polacek et al., 1997; Yeh et al., 1997; Plenz et al., 2004; Wang et al., 2005; Chadjichristos et al., 2006; Li et al., 2012b). Although treatment with the cholesterol-lowering lovastatin or the angiotensin-converting enzyme inhibitor ramipril was shown to decrease both Cx43 expression and neointima formation after balloon injury, a causal link remained to be proven (Wang et al., 2005; Li et al., 2012b). Thus, hypercholesterolemic Cx43+/−Ldlr−/− mice and Cx43+/+Ldlr−/− controls were subjected to carotid balloon distension injury in vivo, which induced marked endothelial denudation and activation of medial smooth muscle cells (Chadjichristos et al., 2006). This genetic reduction in Cx43 expression was found to limit neointima formation after the acute vascular injury by decreasing the inflammatory response as well as reducing smooth muscle cell migration and proliferation. Moreover, endothelial repair was enhanced in mice with reduced Cx43 (Chadjichristos et al., 2006). This suggests that Cx43 might be an attractive target for local delivery strategies aimed at reducing restenosis. Interestingly, effective knock-down of Cx43 was recently achieved with a lentiviral vector expressing Cx43-targeting shRNA and also resulted in an attenuation of neointima formation after balloon injury in rats (Han et al., 2015). Surprisingly, smooth muscle-targeted knockout of Cx43 enhanced neointima formation in response to carotid wire or occlusion injury (Liao et al., 2007). Thus, the level to which Cx43 is reduced as well as the progression of the disease may be of crucial importance for the final outcome. Additional research will be needed before moving on to a translational setting of targeting Cx43 for the reduction of restenosis.
V. Connexins in Cardiac Disease
The heart mainly expresses the connexin isoforms Cx31.9, Cx37, Cx40, Cx43, and Cx45, whereby Cx43 is the most abundant isoform and is mainly found in ventricular cardiomyocytes (Severs et al., 2008) but is also expressed in fibroblasts (McArthur et al., 2015) and stem cells (Lu et al., 2012b). Other isoforms like Cx40 are expressed in the atria of the heart; Cx31 and Cx45 in the conductance system; and Cx37, Cx40, and Cx43 in the coronary circulation. The expression of Cx43 in cardiomyocytes is affected by many factors, including factors released by activation of the neurohumoral system (Salameh et al., 2013) or immune system (Zhang et al., 2016a) as well as by microRNAs (miRNA; further discussed in section V.B.1). As introduced earlier, connexins have a fast protein turnover with a half-life in the order of 1–2 hours in cardiomyocytes, necessitating specific requirements in the organization of the connexin turnover cycle (see section II.B). As a result, the high level of forward connexin trafficking to the sarcolemma needs to be balanced by a well-organized degradation machinery that encompasses multiple proteolytic (Beardslee et al., 1998) and lysosomal (Laing et al., 1997) pathways. Forward trafficking of connexins to the sarcolemma is a crucial component of the connexin life cycle in cardiomyocytes, and in the human heart it has been demonstrated that four truncated connexin isoforms with a molecular weight of ∼20 kDa play a central role because their ablation arrests trafficking of full-length connexin (Smyth and Shaw, 2013). At the intercalated discs, Cx43 interacts with various scaffolding proteins including ZO-1 and others (see section II.B.3). Transmembrane protein 65 (Tmem65) is another scaffolding protein, which functionally regulates Cx43 and thereby cellular coupling (Sharma et al., 2015). The end-station of forward trafficking is the perinexus, a zone at the periphery of the nexus that contains the GJ plaques at the intercalated disks (IDs) (Rhett et al., 2011; Rhett and Gourdie, 2012). In the perinexus, Cx43 arrives as HCs, which then come loose of their ZO-1 binding to become incorporated into the GJs. The perinexus as a defined zone of HC residency is likely to be the site where HC opening may occur. HC opening can be triggered by lowering [Ca2+]e, increasing [Ca2+]i, connexin dephosphorylation, metabolic inhibition, or hyperosmolar conditions, as demonstrated in isolated ventricular cardiomyocytes by dye uptake and electrophysiological methods (Kondo et al., 2000; John et al., 2003; Wang et al., 2012a). Within cardiomyocytes, Cx43 is located also in subsarcolemmal mitochondria (see below) and the CT of Cx43 translocates into the nucleus and inhibits cell proliferation (Dang et al., 2003; Zhao et al., 2015). Cardiac connexins also have extensive non-channel functions that in many cases link to interactions of the Cx43 CT with various scaffolding and signaling proteins, forming a connexin interactome network or “connexome.” Cx43 interacts with proteins related to various biologic processes such as cell cycle, metabolism, signaling, and trafficking (see also section II.C). Importantly, the interactome of Cx43 is differentially modulated in diseased hearts (for review, see Martins-Marques et al., 2015a). Work of the Delmar group has demonstrated that the Cx43 CT is involved in interactions that influence capture of the microtubule plus end at the ID and thereby facilitate sarcolemmal delivery of Na+ channels at the cell end (Agullo-Pascual et al., 2014b; reviewed in Leo-Macias et al., 2016). Thus, alterations in connexin trafficking are likely to influence cell-cell coupling as well as electrical excitability, thereby leading to arrhythmia. Along this line, the Cx43 interactome partner plakophilin-2, which associates with Cx43 at the ID has been demonstrated to be involved in arrhythmias associated with the Brugada syndrome in genetically predisposed patients (Cerrone et al., 2014; Agullo-Pascual et al., 2014a). Figure 7 summarizes some of the proposed roles of GJs and HCs in cardiac ischemia-reperfusion injury.
Roles of Cx43 in cardiac ischemia-reperfusion. Gap junctions close under ischemic conditions (“healing over” caused by the low pH and elevated [Ca2+]i) but substantial coupling may persist after ischemia-reperfusion (Ruiz-Meana et al., 2001). Open GJs may act beneficially by supplying essential metabolites to neighboring cells but may also spread injury signals, causing cell death propagation (reviewed in Decrock et al., 2009b; Michela et al., 2015). GJs have been implicated in the spreading of hypercontracture necrosis in a process mediated by Na+ flux through GJs (reviewed in García-Dorado et al., 2004). Uncontrolled hemichannel (HC) opening facilitates ionic fluxes that may lead to cell swelling (Wang et al., 2013c). Cx43 is also present in mitochondria where they are involved in the signaling cascade of ischemic preconditioning, conferring cardioprotective effects (reviewed in Schulz et al., 2007; Miura et al., 2010; Schulz et al., 2015). Mitochondrial Cx43 has been demonstrated to form hexameric structures involved in inner mitochondrial membrane K+ fluxes, pointing to functional HCs (Miro-Casas et al., 2009).
A. Cx43 in Mitochondria
Cx43 is detected in cardiomyocyte mitochondria from mouse, rat, porcine, and human origin using antibody-dependent and -independent techniques (Boengler et al., 2005; Miro-Casas et al., 2009; Jovic et al., 2012). The analysis of subsarcolemmal and interfibrillar mitochondria for the presence of Cx43 shows that Cx43 is almost exclusively localized in subsarcolemmal mitochondria (Boengler et al., 2009; Sun et al., 2015b). Cx43 is encoded in the nuclear genome and imported into the mitochondria in a heat shock protein 90, translocase of the outer membrane 20-dependent fashion. Cx43 is located at the inner membrane of subsarcolemmal mitochondria, with its carboxy terminus oriented toward the intermembrane space (Rodríguez-Sinovas et al., 2006), but may also be present at the outer mitochondrial membrane regulating the release of some intermembrane space proteins (Goubaeva et al., 2007).
In the inner membrane of subsarcolemmal mitochondria, Cx43 interacts with the apoptosis-inducing factor, the β-subunit of the electron-transfer protein (Denuc et al., 2016) and the ATP-sensitive potassium channel subunit Kir6.1 (Waza et al., 2014). The amount of mitochondrial Cx43 declines in cardiomyocytes after the activation of the N-methyl-d-aspartate receptor 1 by the enhanced mitochondrial translocation of the matrix-metalloproteinase 9 (Tyagi et al., 2010). In isolated mitochondria of Cx43-deficient mice hearts, the reduced mitochondrial Cx43 content is associated with a switch of the mitochondrial NOS isoform and the decrease in mitochondrial nitric oxide formation (Kirca et al., 2015).
In isolated subsarcolemmal mitochondria, chemical cross-linking generates Cx43-dependent protein-complexes at a molecular weight corresponding to that of GJ-enriched membranes, possibly representing Cx43 hexamers. Mitochondrial uptake of the Cx43-formed channel permeable dye Lucifer yellow, which is inhibited by the chemically unrelated Cx43-formed channel blockers carbenoxolone and heptanol, suggests a functional Cx43-formed channel within the inner membrane of subsarcolemmal mitochondria (Miro-Casas et al., 2009; Soetkamp et al., 2014). The finding that the administration of the Cx43 HC-specific peptide Gap19 as well as the genetic ablation of Cx43 reduce the potassium influx into the mitochondrial matrix further strengthens the hypothesis that Cx43-formed channels are present in the inner membrane of subsarcolemmal mitochondria (Miro-Casas et al., 2009; Soetkamp et al., 2014). Subsarcolemmal mitochondria are more responsive than interfibrillar mitochondria to fibroblast growth factor-2–triggered protection from calcium-induced permeability transition pore opening by a mitochondrial Cx43 channel-mediated pathway (Srisakuldee et al., 2014).
Apart from ion fluxes, both a pharmacological inhibition and a genetic ablation of Cx43 reduce mitochondrial complex 1-mediated oxygen consumption and ATP production, whereas complex 2-mediated respiration is not affected. Possibly, Cx43 interacts with proteins of complex 1 of the electron transport chain (Boengler et al., 2012; Denuc et al., 2016). Mitochondria are central to ROS formation and a relationship between Cx43 expression and ROS formation has been established (Matsuyama and Kawahara, 2011; Denuc et al., 2016), whereby Cx43-formed channel inhibition or Cx43 downregulation decrease the amounts of ROS formation.
It is not yet known whether all Cx43 phosphorylation sites described in section II.F are also phosphorylated in mitochondrial Cx43. However, phosphorylation of mitochondrial Cx43 is found at Ser-262 in rat subsarcolemmal mitochondria (Srisakuldee et al., 2014), whereas Cx43 phosphorylation at Ser-368 is detected in mouse, rat, and porcine subsarcolemmal mitochondria (Boengler et al., 2011; Srisakuldee et al., 2014; Shan et al., 2015). Phosphorylation of Cx43 is not necessarily achieved outside the mitochondria, since PKC is present in mitochondria and the stimulation of such mitochondrial PKC leads to Cx43 phosphorylation at Ser-262 and Ser-368 in rat subsarcolemmal mitochondria (Srisakuldee et al., 2014). Activation of mitochondrial PKC renders mitochondria more tolerant toward Ca2+ overload (Srisakuldee et al., 2014).
In addition to the well documented protective role of Cx43 in cardiac subsarcolemmal mitochondria, recent evidence indicates that mitochondrial Cx43 may also contribute to cardiac injury and cell death. Gadicherla et al. (2017) demonstrated that Cx43 forms functional hemichannels in cardiac subsarcolemmal mitochondria, which contribute to Ca2+ entry and trigger permeability transition and cell injury/cell death (Gadicherla et al., 2017). Compared to Gap26 and Gap19, RRNYRRNY peptide appeared to most active in inhibiting mitochondrial Cx43 hemichannel activity. The RRNYRRNY peptide also strongly reduced the infarct size in ex vivo cardiac ischemia-reperfusion studies (Gadicherla et al., 2017). Thus, Cx43 in cardiac subsarcolemmal mitochondria may protect in the context of ischemic preconditioning, but in the absence of preconditioning, it contributes to cardiac injury provoked by ischemia.
B. Cx43 and Risk Factors of Cardiovascular Diseases
1. Age and Sex.
In the promotor region of the Cx43 gene, a series of half-palindromic estrogen response elements is present. In HeLa cells transfected with a luciferase-connexin43 promoter fusion construct, an upregulation of luciferase expression by estrogen occurs (Yu et al., 1994). Although Cx43 mRNA levels in ventricular tissue from neonatal male and female littermates are similar, the Cx43 mRNA levels are higher in adult female compared with age-matched male hearts (Rosenkranz-Weiss et al., 1994). Interestingly, phenylephrine treatment increases Cx43 expression only in female cardiomyocytes (Stauffer et al., 2011). Also on the protein level, expression of Cx43 is markedly lower in males of both normotensive and hypertensive rats compared with female rats (Knezl et al., 2008). The difference in Cx43 expression between males and females is predominantly seen at the level of GJs (Thomas et al., 2011). Multiple reports have documented that Cx43 protein expression declines with age in the hearts of hamsters (Chen and Jones, 2000), guinea pigs (Jones et al., 2004; Jones and Lancaster, 2015), mouse (Bonda et al., 2016), rats (Watanabe et al., 2004; Lancaster et al., 2011), and rabbits (Yan et al., 2013). Such reduced expression with age is associated with enhanced lateralization of the protein as observed in rat and rabbi