Skip Navigation

Cardiovascular Research 2004 61(3):386-401; doi:10.1016/j.cardiores.2003.11.039
© 2004 by European Society of Cardiology
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by García-Dorado, D.
Right arrow Articles by Ruiz-Meana, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by García-Dorado, D.
Right arrow Articles by Ruiz-Meana, M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Copyright © 2003, European Society of Cardiology

Gap junction-mediated spread of cell injury and death during myocardial ischemia–reperfusion

David García-Dorado*, Antonio Rodríguez-Sinovas and Marisol Ruiz-Meana

Laboratorio de Cardiología Experimental, Servicio de Cardiología, Hospital Universitario Valle de Hebron, Barcelona, Spain

* Corresponding author. Laboratorio de Investigación Cardiovascular, Hospitals Vall d'Hebron, Pg. Vall d'Hebron 119–129, 08035 Barcelona, Spain. Tel.: +34-93-4894038; fax: +34-93-4894032. dgdorado{at}vhebron.net

Received 3 October 2003; revised 19 November 2003; accepted 25 November 2003


   Abstract
Top
 Abstract
1. Structure and function...
 2. GJMIC in the...
 3. GJMIC during myocardial...
 4. Gap junction-mediated spread...
 5. GJ and ischemic...
 6. Conclusion and therapeutic...
 References
 
Gap junction-mediated intercellular communication (GJMIC) has been known for a long time to be essential in propagation of electrical impulse in the heart, and the contribution of altered GJMIC to the genesis of arrhythmias has been extensively investigated. However, although it is well known that GJMIC allows the exchange of biologically important molecules between adjacent cells, the pathophysiological significance of such chemical coupling during myocardial ischemia and reperfusion is much less known. It has been solidly established that ischemia impairs GJMIC and eventually leads to electrical uncoupling, but recent studies suggest that GJMIC may still allow synchronization of the onset of ischemic rigor contracture and of the progression of ischemic injury beyond rigor onset. During reperfusion, GJMIC has been shown to mediate cell-to-cell propagation of hypercontracture and cell death, and there is evidence that this phenomenon explains the continuity of areas of contraction band necrosis and contributes to final infarct size. Finally, there is increasing evidence that GJ or their protein components are involved in the genesis of the protective effect of ischemic preconditioning, although probably through mechanisms independent from modulator of GJMIC. GJ play an important role in the pathphysiology of cell injury and death during myocardial ischemia–reperfusion and are potential targets for new cardioprotective therapeutic strategies.

KEYWORDS Myocardial infarction; Ischemic heart disease; Cardiomyocytes

Gap junction-mediated intercellular communication (GJMIC) has been known for a long time to be essential in the propagation of electrical impulse in the heart. The effects of abnormalities in the function, amount, or distribution of gap junctions (GJ) on cell-to-cell electrical coupling, myocardial electrical behavior, and in the genesis of arrhythmias have been extensively investigated. However, although it is well known that GJ allow the diffusion of biologically important molecules, the pathophysiological significance of such chemical coupling has been much less investigated. This review analyzes recent evidence indicating that GJMIC may play a role in the progression and spread of cell injury and death during myocardial ischemia–reperfusion.


   1. Structure and function of GJ
Top
 Abstract
 1. Structure and function...
2. GJMIC in the...
 3. GJMIC during myocardial...
 4. Gap junction-mediated spread...
 5. GJ and ischemic...
 6. Conclusion and therapeutic...
 References
 
1.1 Structure of GJ
GJ are specialized membrane areas containing a tightly packed array of channels (junctional channels, JC) that allow the cytoplasm of two adjacent cells to be directly connected. GJ are ubiquitous in animal cells, have been demonstrated in many phyla from invertebrates and metazoa to chordates, and are generally thought to be absent in plants [1].

The structure of JC has been described in detail elsewhere [1,2]. Each JC is formed by two end-to-end connected hemichannels (also known as connexons) contributed by each of the two adjacent cells. In chordates, hemichannels are hexameric structures formed by six connexins (Cx). Cx have four transmembrane domains with two extracellular loops and with the N-terminal and C-terminal domains in the cytoplasm [1,2]. There is a large multi-gene family of highly homologous connexins with different biophysical properties that are commonly named according to their molecular mass (in kDa). Cells can usually express a variety of Cx, and the six connexins forming a hemichannel may not be identical (for example Cx43 and Cx40), thus forming a heteromeric hemichannel. Although not all combinations are possible [1], hemichannels formed by a particular Cx can dock to hemichannels with a different composition [1,3].

1.2 Permeability and selectivity
The biophysical properties of GJ are incompletely understood. It is now recognized that JC present complex biophysical properties including voltage and chemical gating, permeant selectivity, and rectification [1,4–10]. In addition, the properties of hemichannels are different depending on whether or not they are docked to other hemichannels to form JC [10,11].

The biophysical properties of hemichannels, unitary JC, and GJ have been recently reviewed [1,3–6]. For the purposes of the present review it is useful to summarize here some fundamental concepts.

GJ are permeable to relatively large molecules [1]. Depending on the Cx type, pore diameter ranges between approximately 6.5 and 15 Å [1,12,13], which is wide enough to allow the passage of water, all relevant cations and anions, including Cl–, Na+, K+, and Ca2+ and most second messengers as IP3, cAMP, and cGMP [1,10]. For small molecules JC behave as rather unselective pores. However, as the size of the permeant increases, selectivity becomes more apparent. Different Cx types vary in their permeability to different cytoplasmic molecules [1]. Anion or cation selectivity depends also on Cx type, and for atomic ions ranges from slightly anion selective to highly cation selective [1,5,12–15].

1.3 Gating and regulation
The mechanisms by which JC close or open are not fully elucidated. The available evidence suggest that gating can take place in at least two independent ways [1,4,6,7,9]: a rapid, voltage-driven mechanism that can change the channel conformation between a fully open and a nearly completely closed state within a few ms (voltage gating), and a slower (up to 30 ms) mechanism that brings the channel to complete closure in response to chemical interactions (chemical gating) or to changes in voltage (loop gating). Notably, the voltage dependence of the rapid and slow gating systems may be different [1,4,9]. The regulation of gating is different for each Cx type.

The macroscopic GJ conductivity depends both on the conductivity of open JC and of the number of open JC. It is important to note that regulatory mechanisms may have opposite effects on the average conductivity of open channels (size of the opening) and on the probability of the open state. According to this, a given stimulus may reduce the permeability to large molecules while increasing the macroscopic conductance to small ions and enhancing electrical coupling [16–18]. Thus, the effects of various conditions on GJ cannot be safely summarized as leading to an "increase" or "decrease" in GJMIC but must be evaluated for their effects on permeability, selectivity and electrical coupling.

The modulatory effect of phosphorylation on GJMIC has been the object of recent extensive reviews [19–22]. Phosphorylation alters the probability of the different conductance states as well as intracellular trafficking and assembly of Cx depending on the Cx type, the phosphorylation site and, possibly, the biochemical environment [19].

The number of GJ that connect two cells is determined by the rates of Cx synthesis, assembly, trafficking, and degradation, which in turn are regulated by multiple factors including G proteins [23] and phosphorylation by different protein kinases [20]. Newly synthesized Cx are assembled into hemichannels that are transported to the plasma membrane. Here they are incorporated into the periphery of GJ where they dock to an opposing hemichannel to form a JC [24]. "Old" Cx are removed from the center of GJ by an endocytic process, and degraded by lysosomes and proteasomes. The half-life of Cx is relatively short (less than 2 h for Cx43) [25] so that changes in the rates of synthesis or degradation are rapidly translated into changes in the number of operative JC.

The physiological relevance of the regulation of GJMIC has remained obscure for a long time. Exceptions are the closure of GJ during cell death and cell division. It has become apparent that dying cells are isolated from adjacent cells in order to prevent the spread of injury [26], a phenomenon often referred to as "healing over" [26,27]. Interruption of GJMIC during cell division [28,29] is carried out through mechanisms dependent on the activation of phosphorylation cascades [30]. Moreover, impaired GJMIC has been proposed to contribute to the increased division rate in certain tumor cells [31].

1.4 Pharmacological control
Although many substances have effects on GJMIC, there is a lack of powerful and specific GJ blockers [16,21]. The main characteristics of the most widely used GJ blockers are summarized in Table 1. On the other hand, several antiarrhythmic peptides have been described that specifically increase GJ conductance, such as AAP10 [32–34] or its derivative ZP123 [35].


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

  		Table 1 Main characteristics of GJ inhibitors representative of different chemical groups

 

   2. GJMIC in the heart
Top
 Abstract
 1. Structure and function...
 2. GJMIC in the...
3. GJMIC during myocardial...
 4. Gap junction-mediated spread...
 5. GJ and ischemic...
 6. Conclusion and therapeutic...
 References
 
GJMIC is particularly well developed and functionally relevant in the mammalian heart. It was in the heart where the existence of direct intercellular communication was first proposed in the 1920s [36] and where GJ structures were first demonstrated by electron microscopy studies around 1960 [37]. The heart is formed by different cell types, each of them expressing multiple Cx types [19,38,39]. However, this review will focus on ventricular cardiomyocytes, which express almost exclusively Cx43, only moderate levels of Cx45, and, to a much lower degree, Cx40 [38,39]. Cx43 JC has the widest pore, is the most anion-selective, and has intermediate unitary conductance [1,12–15,40–42]. The most prominent biophysical properties of Cx43, Cx45, and Cx40 are shown in Table 2.


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

  		Table 2 Biophysical properties of Cx43, Cx45 and Cx40

 
Maturation of the myocardium is associated with redistribution of GJ towards the membrane interdigitations in the intercalated disks so that early after birth, GJ are rare outside these structures [43]. Thus, adult ventricular myocardium can be considered to be end-to-end connected.

The best-characterized function of GJMIC in myocardium is to allow rapid propagation of the electrical impulse [44,45]. Under physiological conditions the conductance of myocardial GJ has a large safety factor regarding impulse propagation. Propagation may be maintained after closure of more than 50% of GJ [44,45], although reduction in GJ conductance is associated with an increase in tissue anisotropy that may be arrhythmogenic due to a predominant effect on transverse conduction [44,46,47]. The evidence demonstrating the critical role of GJ in impulse propagation is overwhelming. Nevertheless, according to some authors the electrical field associated with action potential depolarization may induce depolarization of adjacent cells even in the absence of GJ [48].


   3. GJMIC during myocardial ischemia
Top
 Abstract
 1. Structure and function...
 2. GJMIC in the...
 3. GJMIC during myocardial...
4. Gap junction-mediated spread...
 5. GJ and ischemic...
 6. Conclusion and therapeutic...
 References
 
Closure of myocardial GJ during ischemia was proposed by McCallister et al. in 1979 [49]. The hypothesis of healing over in ischemic cardiomyocytes was soon accepted as dogma, supported by two main lines of evidence: first, ischemia induces dramatic changes in the electrical behavior of the myocardium that can be explained by GJ closure [50]; second, prominent cytosolic derangements characteristic of ischemia induce GJ closure when applied to cell systems in vitro [51,52].

3.1 Electrical properties of ischemic myocardium
Analysis of myocardial preparations [50–56] and in vivo studies [57–59] show that ischemia induces abnormalities in electrical tissue resistance and impulse propagation [53–59]. These changes appear after a period of latency in close temporal correlation with the onset of ischemic rigor contracture [54,56] and develop progressively during a relatively prolonged period of time (Fig. 1). Abnormalities can be divided into those affecting the passive electrical properties (i.e. properties of resting myocardium as a passive conductor), and those modifying active electrical properties (i.e. those affecting generation and propagation of the action potential). Passive properties are studied by applying currents that are too small to cause cell depolarization (subumbral stimuli), while active properties are studied by applying excitatory pulses.


Figure 1
View larger version (12K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 Effect of ischemia on electrical coupling in the isolated rabbit papillary muscle. After a period of latency, ischemia induces a progressive increase in cytosolic [Ca2+] (top panel), an abrupt increase in tissue resistance (Rt), and rigor contracture. There is a close temporal relationship between the onset of the three phenomena (from Dekker et al. [54], with permission). This relationship is maintained under different conditions (see Refs. [57,65,120]).

 
Linear electrical resistance of a conductor cable segment can be defined as the relation between the difference in voltage between two points and the intensity of the current between them. However, myocardial tissue is a complex material composed of multiple elements connected according to a complex pattern. In a very simplified way, the myocardial circuit contains three main elements corresponding to the intracellular compartment, the extracellular compartment, and the cell membranes [60]. The intracellular resistance (the resistance that would be found if current were flowing exclusively through cytosol and GJ) and the extracellular resistance (the resistance that would be found if current were flowing exclusively through the extracellular space) are arranged "in parallel", and cell membranes act as a capacitor connecting them [60]. Unfortunately, intracellular and extracellular resistances cannot be in general measured separately. This separate analysis can only be performed indirectly under very precise conditions such as those described by Kleber et al. [53]. These authors compared the responses to subumbral and excitatory pulses in the cable-like rabbit papillary muscle.

The results of studies performed in this model are rather consistent. After an initial period during which intracellular resistance does not change, ischemia induces an abrupt increase in both total and intracellular tissue resistances that is interpreted as the consequence of a reduction in GJ conductivity (uncoupling), and a reduction in the velocity of impulse propagation (conduction velocity) [53,54,56]. The onset of these changes seems to be linked to the development of ischemic rigor contracture, a phenomenon that marks critical levels of ATP and onset of the rise in the cytosolic Ca2+ concentration (Fig. 1) [54,55]. Increased cytosolic Ca2+ concentrations, reduced ATP concentration, and acidification are known to close GJ channels [52,61–63]. Moreover, simultaneous analysis of cytosolic [Ca2+] by ratiofluorescence methods has shown that the onset of uncoupling is linked (with a delay of approximately 2 min) to the onset of the rise in [Ca2+] [54]. Changes in the myocardial electrical properties progress rapidly, and 10–20 min after development of rigor contracture, maximal tissue resistance and complete conduction block are reached [53–56].

A more widely applicable approach for assessment of passive electrical properties is the analysis of myocardial impedance. When the current applied to a circuit is not constant but follows a curve along time (alternating current), voltage follows a curve with the same shape. In the absence of capacitors, the two curves would be synchronous. However, capacitors can accumulate current and release it into the circuit, and their presence causes a certain delay or shift in the curve of voltage with respect to the curve of current. Since the myocardium contains capacitor components (basically, cell membranes), its electrical properties cannot be adequately characterized by its linear resistance, but by its electrical impedance. This parameter includes two main components: resistance (relationship between current intensity applied and voltage measured) and capacitance (delay between the current and voltage curves). For analysis of myocardial impedance, the measured voltage follows the same wave pattern as that of the current applied and the delay is measured as the phase angle (a delay equivalent to the full wavelength corresponds to a 360° phase angle shift). Resistance and phase angle shift are dependent on the frequency of the applied voltage wave, and previous studies have shown that some frequencies allow a better discrimination between ischemic and control myocardium [64]. However, it should be stressed that impedance analysis does not allow separate measurement of intracellular and extracellular resistance.

Studies using analysis of myocardial electrical impedance have also provided consistent results [57–59,65]. In in situ ischemic porcine myocardium, resistivity and phase angle shift show a rapid initial change immediately after coronary occlusion that is thought to reflect acute changes in the extracellular space and is followed by a period of stabilization during which impedance changes slowly [57–59,65]. This period is interrupted by an abrupt change in the resistivity and phase angle curves that is closely related to the onset of rigor contracture [57–59,65]. This abrupt change is interpreted as the consequence of GJ closure and electrical uncoupling [53,54,56] and is followed by a period of 90–120 min during which resistivity and phase angle shift rise steadily [57–59,65].

The changes induced by ischemia in myocardial electrical properties have been interpreted as evidence of the healing over phenomenon in ischemic myocardium and as proof that rigor onset is rapidly followed by GJ closure and chemical and electrical isolation of cardiomyocytes [50,51,54]. This concept has found additional support by direct observations of GJ (macroscopic) and single channel conductances under conditions that reproduce some of the critical derangements induced by ischemia.

3.2 Analyses of GJ conductance in cells and cell expression systems
GJ conductance can be directly measured by means of the double whole-cell patch clamp. This technique requires that two coupled cells are simultaneously voltage-clamped to a common potential so that there is no voltage gradient across GJ. In essence, the potential of one of the two cells is increased in a step while the potential in the other cell is kept constant. This creates a voltage gradient across GJ that induces a current. This current equals the current that has to be supplied to the second cell to hold its potential constant, and can be thus measured. From the voltage and the intensity, the GJ resistance (or its inverse, the conductance) is calculated [1]. By altering the conditions in such a way that the number of functioning JC is sufficiently small, it is possible to measure voltage and intensity across a single channel, and unitary JC conductance can be calculated [1].

By using this technique in pairs of connected cardiomyocytes, in coupled pairs of oocytes, and in other expression systems, previous studies have shown that conductance of Cx43 channels is reduced by increased cytosolic concentrations of H+ [61] and Ca2+ [52,62], by depletion of ATP [63], and by accumulation of amphipathic metabolites [66], abnormalities that concur in ischemic myocardium.

Gating by H+ concentration has been particularly well characterized [1,61,67] and has been proposed as a model of chemical gating [68]. Acidosis reduces conductance of Cx43 in a dose-dependent manner and at pH 6.4 the channel is closed [61]. Mutation experiments suggest that H+ regulation operates on a sensor located in the C-terminal cytosolic domain of Cx43 through a direct proton–protein interaction. This interaction induces a conformational change in this segment that occludes the cytosolic end of the pore by a "ball and chain" mechanism [1,68], which has been proposed as the molecular basis of chemical gating [68]. An alternative hypothesis proposes that most chemical mediators close GJ through cytosolic components, such as calmodulin, that would physically occlude the channel (the "cork" hypothesis) [1,52]. The mechanism of gating by ATP and Ca2+ concentration remains obscure. However, moderate increases in [Ca2+] have been shown to cause uncoupling [52,61,62]. The sensitivity to Ca2+ seems to be increased at low pH in a kind of positive interaction [67,69].

3.3 Challenging the healing over dogma
Although the data demonstrating that ischemia reduces electrical coupling are extremely solid, the generally accepted view that GJMIC is abolished in ischemic myocardium rests on weaker evidence, and the persistence of residual chemical communication through GJ in severely damaged ischemic myocardium cannot be excluded based on present data.

Indeed, the alterations induced by ischemia on myocardial electrical properties are not incompatible with persistent chemical coupling. First, the slow progression of electrical abnormalities could reflect, in part, a progressive decrease of GJ conductivity and the persistence of residual electrical coupling in severely ischemic cells over prolonged periods of time. Moreover, the complete blockade of impulse propagation may occur before GJ conductivity reaches zero [53,57,65,70] due to depressed excitability [71,72]. The dissociation between GJ conductance and conduction blockade is well illustrated by the fact that under certain circumstances, such as current-to-load mismatch, conduction block may occur in the presence of normal GJ conductance [71–73].

On the other hand, studies in cell pairs or expression systems conclusively showing GJ closure in response to single modifications such as acidosis or increased Ca2+ cannot be easily extrapolated to ischemic myocardial tissue, a situation associated with a myriad of cytosolic alterations with varying and even opposite effects on GJMIC. There are complex and contradictory effects of real ischemia on GJMIC; these are well exemplified by the effect of ischemia on phosphorylation-mediated GJ regulation. Ischemia causes initial activation of different protein kinases, including MAP kinases and protein kinase C (PKC-{varepsilon}) with different effects on GJMIC. In Cx43 GJ, MAPK-dependent phosphorylation reduces GJMIC [20,74], whereas the effects of PKC activation are more controversial. Most of the studies suggest that PKC-dependent phosphorylation shifts GJ conductance to a lower value (narrower pore diameter) [20,74]. Treatment with the PKC activator TPA in different cell types caused a reduction in permeability, which correlates with the predominance of low-conductance single channel events [18,75]. Similarly, the reduction in GJ permeability observed after treatment with fibroblast growth factor-2 (FGF-2) was dependent on phosphorylation of Cx43 mediated by PKC-{varepsilon} [76]. However, these effects seem to depend on the active isoform, as PKC-{alpha} stimulation may enhance GJ conductance [34]. Later on, reduced ATP levels and phosphorylation potential may result in dephosphorylation of Cx43, which tends to increase GJMIC [77]. In addition, ischemia markedly reduces soluble and particulate guanylyl cyclases and cytosolic cGMP [78,79], which should increase GJMIC since PKG-mediated phosphorylation reduces the open probability of Cx43 channels [74]. After prolonged, sustained ischemia, the altered turnover of Cx43 may contribute to reduction of the number of JC and coupling [20]. The importance of the differences between in vitro cell pairs and ischemic myocardium is illustrated by the fact that exposure to pH 6.4 under normal extracellular [Ca2+] completely abolishes Cx43 conductivity [61,80], whereas no reduction in cell coupling is observed during the initial period of myocardial ischemia despite a severe acidosis [53] (Fig. 2).


Figure 2
View larger version (21K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 Effect of acidosis on cell coupling under in vitro and in vivo conditions. (A) In Xenopus oocytes expressing Cx43, GJ conductance is reduced by acidosis in a dose-dependent manner (closed circles). The effect of acidosis is markedly inhibited by injection of a polypeptide corresponding to amino acids 241 to 382 of Cx43 (carboxyl terminal domain of C Cx43) (open circles). Modified from: Calero et al. [144], with permission. (B) Time course of intracellular acidosis in rat hearts submitted to global ischemia, as measured by nuclear magnetic resonance. After only 10 min of ischemia intracellular acidosis is pronounced. Modified from Rehring et al. [145], with permission. (C) Effect of ischemia on myocardial electrical impedance (phase angle mean values) in pig myocardium during 48 min coronary occlusion and 30 min reperfusion as an index of changes in junctional conductance. The abrupt change associated with the onset of electrical uncoupling (arrow) occurs after only approximately 20 min of coronary occlusion despite the much earlier occurrence of severe acidosis. Modified from Rodriguez-Sinovas et al. [57], with permission.

 
Recent studies have suggested the persistence of GJMIC in ischemic cardiomyocytes (Fig. 3) [81]. In end-to-end connected cardiac myocytes submitted to metabolic inhibition, the development of rigor contracture in one of the cells was immediately followed by the development of rigor contracture in the adjacent cell. This close temporal association was in sharp contrast with the large variability observed in non-connected cells and was sensitive to the GJ uncoupler 18{alpha}-glycyrrhetinic acid. The curves showing the rise in cytosolic Ca2+ ([Ca2+]i) observed after rigor onset were identical in connected cells, and persistence of GJMIC was demonstrated by transfer of the GJ-permeable dye Lucifer Yellow (LY) up to 20 min after rigor development. LY transfer was also documented 30 min after the onset of rigor contracture in isolated rat hearts submitted to non-flow global ischemia (Fig. 3) [81]. These results suggest that GJMIC may allow cell-to-cell propagation of rigor contracture and equalization of Ca2+ overload in ischemic myocardium. Although the mechanism of propagation of rigor contracture is not clear, it can be speculated that cells developing rigor and consuming ATP in an accelerated way may "steal" GJ-permeable ATP from adjacent cells, decreasing their ATP levels to the critical values at which rigor contracture develops.


Figure 3
View larger version (66K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 Synchronization of ischemic rigor contracture and persistence of dye coupling. Top: Serial fura-2 340/380 ratiofluorescence images of adult rat cardiomyocytes (C1, C2 and C3) submitted to simulated ischemia (metabolic inhibition at pH 6.4). The arrow indicates the intercalated disk connecting C1 and C2. In the middle frame one of the two connected cells (C1) has developed rigor contracture and its cytosolic Ca2+ concentration has increased. In the right frame both cells (C1 and C2) are contractured. The curves of fura-2 340/380 ratio in the two connected cells C1 and C2 are synchronized and independent from the curve in a separated cell (C3). Middle: Serial fluorescent images (A–F) demonstrating the transfer of the GJ-permeable fluorescent dye Lucifer Yellow injected 10 min after development of rigor contracture. The graph shows the time course of dye transfer (fluorescence intensity). ID: Intercalated disk. Bottom: Dye transfer in the rat heart during normoxia (A1, B1) and 30 min after development of ischemic rigor contracture (A2, B2). A deep incision in the left ventricular wall allowed exposure (under hypoxic conditions) to two fluorescent dyes: Rhodamine-conjugated Dextran (RD) that cannot diffuse through GJ, and LY. After dye loading, hearts were fixed, frozen, sectioned and examined by confocal laser scanning microscopy. RD(+) area (red) corresponds to cells with disrupted sarcolemma along the border of the myocardial incision. LY(+) area (green) marks intercellular dye diffusion through GJ. The area of diffusion of LY in hearts submitted to 45 min of ischemia (B2) was similar to that observed in normoxic controls (B1). Analysis at higher magnification shows intercellular diffusion of LY (B3). Modified from Ruiz-Meana et al. [81], with permission.

 
These results in cardiomyocytes are in agreement with observations demonstrating the persistence of GJ communication during ischemic conditions in tissues other than myocardium. Cotrina et al. [82] and Contreras et al. [83] found that dye coupling between cultured astrocytes was progressively reduced but never abolished during 2 h of metabolic inhibition and that chemical communication through GJ occurred up to the terminal loss of membrane integrity. Interestingly, this was not modified when metabolic inhibition was performed at pH 6.0 [82]. Chemical coupling in astrocytes was corroborated in hippocampal slices during ischemic conditions [82]. Similar results [26] were obtained in neocortices of rats after applying the technique of fluorescent recovery after photobleaching. Although a decrease in fluorescence recovery (reflecting a decreased diffusion through GJ) was observed, the authors found that astrocytic GJ remain open in the anoxic brain.


   4. Gap junction-mediated spread of cell death during reperfusion
Top
 Abstract
 1. Structure and function...
 2. GJMIC in the...
 3. GJMIC during myocardial...
 4. Gap junction-mediated spread...
5. GJ and ischemic...
 6. Conclusion and therapeutic...
 References
 
Early reperfusion (that resulting in salvage of a part of the area at risk) is associated with the rapid re-energizing and disappearance of the ischemic derangements responsible for reduced GJMIC, including elevated [Na+]i and [Ca2+]i, intracellular acidosis, membrane depolarization, and altered extracellular composition. Impedance analysis has demonstrated that this recovery is paralleled by a rapid normalization of myocardial electrical resistivity (Fig. 2C).

The hypothesis of cell-to-cell spread of cell death in reperfused myocardium was proposed to explain the spatial continuity of the areas of contraction band necrosis in myocardium submitted to ischemia–reperfusion [84,85]. When myocardial ischemia is brief enough to cause the death of only a part of the myocytes within the myocardium at risk (severely ischemic), cell death occurs almost exclusively during the first minutes of reperfusion in the form of contraction band necrosis [86,87], a histological pattern that reflects hypercontracture and sarcolemmal rupture of cardiomyocytes [88,89]. Strikingly, hypercontracted, dead cardiomyocytes are not scattered across reperfused myocardium, but are invariably connected to other dead myocytes within well-delimited areas of contraction band necrosis, often with irregular geometry [85]. This pattern cannot be explained as a consequence of microvascular or collateral distribution or other structural patterns, and computer simulation studies indicated that it is due to some kind of cell-to-cell interaction [84].

Hypercontracture is associated with dramatic changes in cytosolic composition and cell geometry, and is likely to cause physical and chemical interactions between adjacent cells. This interaction was demonstrated in end-to-end connected pairs of freshly isolated rat cardiomyocytes submitted to simulated ischemia. Reoxygenation-induced hypercontracture of one of the cells was immediately followed by hypercontracture of its pair [90], which was in sharp contrast to the difference in the time course of hypercontracture in pairs of cells that were close but not in physical contact (Fig. 4) [90]. That it was propagation instead of simple temporal coincidence was demonstrated by inducing hypercontracture in one of the two cells of end-to-end connected pairs that had survived reoxygenation without developing hypercontracture (Fig. 4C). Hypercontracture induced by microinjection of extracellular medium propagated to the adjacent cell in the same manner as spontaneous reoxygenation-induced hypercontracture [90,91]. Microinjection of extracellular media was proposed as a relevant model to investigate the propagation of hypercontracture in myocardial tissue where reperfusion-induced hypercontracture is associated with sarcolemmal rupture and immediate, massive enzyme release [90].


Figure 4
View larger version (76K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4 GJ-mediated propagation of reoxygenation-induced hypercontracture. In rigor-shortened cardiomyocytes (A), reoxygenation-induced hypercontracture was synchronous (time from reoxygenation to hypercontracture was near identical) in connected cell pairs (triangles), but was highly variable in non connected cells (circles) (B). Induction of hypercontracture in a reoxygenated, non-hypercontracted cell by microinjection of extracellular buffer (white arrow) progressed in less than 16 s to the adjacent cell (C). Injection of extracellular buffer with Lucifer Yellow resulted in dye diffusion and propagation of hypercontracture (D) that were prevented by heptanol (E). The black arrows indicate the intercalated disk. (A), (B), (D) and (E) are modified from Garcia-Dorado et al. [90], with permission. (C) modified from Garcia-Dorado et al. [85], with permission.

 
The propagation of hypercontracture was shown to be dependent on GJMIC in experiments in which Lucifer Yellow was added to the extracellular medium that was microinjected into one of two end-to-end connected cardiomyocytes in the presence or in the absence of the GJ uncoupler heptanol [90] (Fig. 4E). GJ closure, demonstrated by absence of diffusion of fluorescence to the adjacent cell, was invariably associated with failure of propagation of hypercontracture [90]. Cell-to-cell propagation of hypercontracture was dependent on induction of a rapid increase in [Ca2+]i in the adjacent cell (Fig. 5A and B). However, this increase could not be explained by the direct passage of Ca2+ through GJ, since it was dependent on the presence of Ca2+ in the extracellular media (Fig. 5C and D): in the absence of extracellular Ca2+, microinjection of Ca2+-containing buffer and Lucifer Yellow resulted in dye diffusion, demonstrating coupling, without propagation of hypercontracture [91]. Instead, the increase in [Ca2+]i in the second cell appeared to be caused by the passage of Na+ from the hypercontracting cell and subsequent reverse-mode Na/Ca exchange [91], as demonstrated by the inhibition of the [Ca2+]i rise and hypercontracture in the adjacent cell in the presence of the Na/Ca exchange blocker KB-R7943 (Fig. 5E and F). Microinjection of Na+ and EGTA to prevent the [Ca2+] rise did not cause hypercontracture of the microinjected cell but was followed by hypercontracture of the adjacent cell, further demonstrating that the passage of Na+ through GJ was responsible of the propagation of hypercontracture [91]. This surprising result can be explained by the larger increase in Na+ entry when the sarcolemma is ruptured and cells are exposed to extracellular media (containing 100 fold more Na+ than Ca2+) as compared to the increase in [Ca2+], by the higher permeability of GJ to Na+ [92,93], and by the high capacity of the sarcolemmal Na/Ca exchanger.


Figure 5
View larger version (70K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5 Role of Na+ transfer and subsequent Na/Ca exchange in GJ-mediated propagation of hypercontracture. (A) Fura-2 340/380 ratiofluorescence imaging of two end-to-end connected and a separate cardiomyocyte. Injection of extracellular buffer in one of the end-to-end connected cardiomyocytes results in a rapid increase in cytosolic Ca2+ concentration and hypercontracture that rapidly propagate to the adjacent cell. The graph (B) shows the changes in cell length in the injected (open circles) and the adjacent cell (closed circles). The asterisk indicates the location of the intercalated disk. The arrow indicates the injection site. (C and D) When injection of extracellular medium with LY is performed in the absence of extracellular Ca2+, hypercontracture does not propagate to the adjacent cell despite the persistence of dye coupling, indicating that Ca2+ enters the adjacent cell from the extracellular space. (E and F) Absence of propagation of hypercontracture despite dye coupling in the presence of Na/Ca blocker KB-R7943. Modified from Ruiz-Meana et al. [91], with permission.

 
Cell-to-cell propagation of hypercontracture raises the question of why this phenomenon does not result in the spread of cell death to the whole area at risk or even beyond. A potential explanation could be that propagation is arrested when a less injured cell is able to handle Na+ and Ca2+ influx efficiently enough so as to prevent severe Ca2+ overload. In addition, Na+ and Ca2+ overload cannot spread when less fragile cells maintain membrane integrity despite hypercontracture.

Studies in isolated rat hearts and in situ pig hearts indicate that GJ-mediated spread of cell death contributes to the final extent of necrosis after myocardial ischemia-reperfusion. Addition of the GJ blocker heptanol at concentrations that do not prevent hypercontracture in single isolated cardiomyocytes has a marked protective effect when applied at the time of reperfusion [90]. Moreover, reperfusion treatment with heptanol modifies infarct geometry, increasing the fragmentation of the areas of contraction band necrosis in agreement with the proposed mechanism of action of interference with cell-to-cell spread of necrosis [90]. Interestingly, a similar effect on infarct geometry was induced by treatment with the contractile inhibitor 2-3-butendionemonoxime (BDM), a compound that has turned out to be a potent inhibitor of GJMIC [94].

The protective effect of GJ inhibitors against reperfusion-induced cell death has been confirmed by other observations. Studies analyzing the effect of other GJ uncouplers (halothane) when administered during reoxygenation or reperfusion in the in situ rabbit heart [95] or the isolated rat heart [96,97] have consistently documented a reduction of necrosis (enzyme release or infarct size). The recent observation that four structurally unrelated GJ inhibitors with different mechanisms of action (heptanol, 18{alpha}-glycyrrhetinic acid, halothane, and palmitoleic acid) share a protective effect against myocardial necrosis when administered at the time of reperfusion [98] supports the hypothesis that the protective effect of heptanol is mediated through its effects on GJMIC. Finally, a recent study analyzing the effect of underexpression of Cx43-deficient mice (Cx43+/–) on cell death secondary to myocardial ischemia also found a protective effect [99]. The lack of viability of Cx43–/– mice has thus far limited the usefulness of Cx43-deficient mice in the study of the role of GJ in the pathophysiology of ischemia-reperfusion injury. This situation could change with the availability of site-conditioned knockout mice [100].

Although there is increasing evidence that most of cell death induced by transient ischemia occurs during the first minutes of reperfusion through contraction band necrosis [84–87,90], the role of apoptosis in myocardial cell death secondary to ischemia–reperfusion is a matter of debate [101], and whether GJMIC may modulate it has not yet been established.

Studies in tissues other than myocardium have also found evidence of GJMIC-mediated spread of cell death under different insults including transient or permanent ischemia. Several studies have shown that GJ coupling during cerebral ischemia can result in an amplification of cell injury [26,102–104]. Similarly, GJ enhance neuronal vulnerability to brain traumatic injury [105] and glutamate cytotoxicity [106]. The ability of GJ to allow spread of cell death (necrotic or apoptotic) to adjacent cells (the bystander effect) may be important in the chemical as well as physical (radiation) therapy of cancer [107–110], but may play an important role in other pathological and physiological conditions [111]. A nice example of the latter is the recent observation of the spatial clustering of apoptotic cell death during development of the retina. Clustering was altered by gap junction inhibitors (octanol or carbenoxolone), and induction of apoptosis by injection of cytochrome c resulted in apoptotic cell death of adjacent cells through a GJ-dependent mechanism [112]. However, the role of GJMIC in cell death is probably complex and may vary depending on the conditions. It has been described that GJMIC may reduce the susceptibility of cells to potentially lethal noxa by allowing "dilution" of the insult and of the secondary cytosolic alterations within a larger cell mass [113–115], a phenomenon known as "the Good Samaritan effect". Thus, some studies have suggested that transfer of some "survival signals" through GJ may play a role in the increased resistance to infarction conferred by ischemic preconditioning in mouse hearts [114]. Cell–cell interaction through GJ has been described to prevent apoptosis of rat neonatal ventricular myocytes [115]. In this regard, free radicals seem to play a secondary role in early lethal reperfusion injury [116,117], but they can induce cardiomyocyte apoptosis [118]. Although studies on the efficacy of scavengers to limit cell death secondary to transient coronary occlusion are extremely controversial, recent data from experiments in genetically modified myocytes suggest that they can have some protective effect [119]. GJ could allow the propagation of antioxidants from non-ischemic or less injured cells, thereby protecting those more severely damaged. The spread of cell injury through gap junctions and the "Good Samaritan effect" are not contradictory phenomena, and it is easily conceivable that GJMIC may result in one or the other phenomenon, depending on the intensity of the insult.


   5. GJ and ischemic preconditioning
Top
 Abstract
 1. Structure and function...
 2. GJMIC in the...
 3. GJMIC during myocardial...
 4. Gap junction-mediated spread...
 5. GJ and ischemic...
6. Conclusion and therapeutic...
 References
 
An intriguing aspect of the role of GJ during myocardial ischemia–reperfusion is their potential involvement in the genesis of ischemic preconditioning. Despite the huge research effort very little is known about the end-effector mechanisms by which preconditioning with brief ischemic episodes may prevent cell death secondary to prolonged ischemia and reperfusion. It was hypothesized that one of these end-effector mechanisms could involve restriction of GJ-mediated spread of injury [120]. PKC{varepsilon} and MAPKs, the most prominent kinases involved in the preconditioning cascade, are known to regulate Cx43 phosphorylation and function [20,74]. Moreover, recent proteomic studies have disclosed that Cx43 participates in the formation of signalling complexes that may play a critical role in the cellular response to PKC{varepsilon} activation [121].

Indeed, there is increasing evidence suggesting an involvement of GJ or Cx43 in ischemic preconditioning. Preconditioned hearts have been described to retain higher levels of Cx43 as compared to non-preconditioned hearts [122], and the administration of the gap junction blocker heptanol before preconditioning ischemia abolished the protective effect in isolated mouse heart [114]. Recent studies have clearly demonstrated that preconditioning attenuates the increase in non-phosphorylated Cx43 occurring during sustained ischemia in pig [123] and rat [124] myocardium. Moreover, underexpression of Cx43 in a transgenic model did not modify the extent of necrosis induced by ischemia–reperfusion but completely abolished preconditioning protection [125].

However, detailed electrophysiological analyses have shown minimal or no effect of previous ischemic preconditioning on active (action potential propagation) or passive (electrical impedance) myocardial electrical properties during prolonged ischemia [65]. More importantly, preconditioning had no effect on the rate of normalization of electrical coupling during reperfusion [65]. These results speak against any contribution of attenuated GJ-mediated spread of cell death in the protective effect of preconditioning and suggest that any relationship between Cx43 and preconditioning should be independent of GJMIC.

A possibility is that the preservation of Cx43 phosphorylation in preconditioned myocardium could contribute to protect myocytes during ischemia by preventing the opening of hemichannels (which would allow water and Ca2+ influx) in response to energy depletion, as observed in rat astrocytes submitted to metabolic inhibition [83]. The role of hemichannels in the genesis of ischemic injury and the effect of ischemic preconditioning in this role remain to be elucidated. Alternatively, Cx43 could be involved in ischemic preconditioning by participating in the signal transduction cascade that elicits protection. Indeed, proteomic analyses have shown that Cx43 participates in the formation of signaling complexes that could play a role in the cellular response to PKC{varepsilon} activation [121]. Finally, very recent studies in astrocytes have shown that the forced expression of Cx43 has a protective effect against cell death secondary to energy depletion (metabolic inhibition) and other types of cell injury [126]. This effect was preserved when GJMIC was prevented by physical separation of cells, by GJ inhibitors, or even when cells expressed mutant, non-functional Cx43 [126], strongly suggesting that the overexpression of Cx43 has protective effects against cell death independent on any effect on GJMIC.


   6. Conclusion and therapeutic implications
Top
 Abstract
 1. Structure and function...
 2. GJMIC in the...
 3. GJMIC during myocardial...
 4. Gap junction-mediated spread...
 5. GJ and ischemic...
 6. Conclusion and therapeutic...
References
 
The main conclusion that can be drawn from previous research on the role of GJ during transient myocardial ischemia is that the pathophysiology of this condition, and the genesis of cell death secondary to ischemia–reperfusion, cannot be properly understood without considering GJMIC. There is solid evidence that GJMIC contributes to the synchronization of the progression of injury across ischemic myocardium and that it allows the spread of hypercontracture-mediated cell death during myocardial reperfusion. GJ and/or Cx43 appear to be involved in the protective effect of preconditioning, but the exact nature of their participation remains obscure.

Application of this knowledge to the design of treatments able to limit myocardial necrosis secondary to ischemia–reperfusion has an obvious potential interest but it is far from being forthcoming. An important limitation is the lack of specific, rapid, reversible, and safe inhibitors of GJMIC. The present status of pharmacological control of GJMIC has been recently reviewed [21]. Polyalcohols (such as heptanol), halothane (and other volatile anesthetics), fatty acids (such as palmitoleic acid), or modifiers of the phosphorylation status, as BDM or 18{alpha}-glycyrrhetinic acid, are not efficacious GJ blockers at low concentrations, and at higher concentrations they have significant effects on other sarcolemmal channels and cell systems.

Another limitation is that in the beating heart suffering coronary occlusion, blockade of GJ in normally perfused myocardium is incompatible with normal electromechanical function. Drugs able to interrupt GJMIC in normal myocardium need to be locally delivered into the area at risk (for example, by intracoronary infusion). A possible way to circumvent this problem comes from observations suggesting a higher susceptibility of ischemic-reperfused myocardium to GJ inhibitors as compared to normal myocardium [98,127]. Even if blockade of GJMIC is circumscribed to the area at risk, changes in impulse propagation within this area could lead to the development of life-threatening arrhythmias [128]. Although uniform GJ blockade within the area at risk should not be expected in theory to be arrhythmogenic, studies involving intracoronary delivery of GJ blockers in large animals are consistent with an increased risk of arrhythmias during regional contractile blockade [129].

For the reasons stated previously, prevention of the spread of cell death by regional GJ blockade during myocardial reperfusion has been shown to be feasible in large animals but has not been described in human patients. However, efforts are being made to develop more selective and safer GJ inhibitors. If adequate GJ inhibitors become available, the increasing use of percutaneous intracoronary interventions to achieve reperfusion in patients with evolving myocardial infarction will offer the possibility to evaluate the balance between the potential risks (mainly arrhythmogenesis) and benefits (infarct limitation) of GJ inhibition in large populations of patients undergoing reperfusion therapy under controlled and safe conditions. Regional GJ blockade could also be tested in reperfusion following surgical ischemia. Since GJ-mediated propagation of hypercontracture appears to be dependent on Ca2+ entry in the adjacent cell via reverse-mode Na/Ca exchange, a possible indirect strategy to prevent it is to block this exchange during the first minutes of reperfusion, an approach that it has been recently shown to be feasible [130].

The rationality behind applying GJ inhibitors during ischemia is less clear, since there is thus far no evidence that GJMIC accelerates or enhances cell injury in ischemic myocardium. In addition, this strategy has the limitations inherent to treatments that need to be applied before coronary occlusion, which generally occurs unexpectedly.

Finally, understanding the participation of GJ and Cx43 in the genesis of ischemic preconditioning could eventually lead to the development of safe and applicable strategies to trigger endogenous myocardial protection, but it appears that a long way still separates us from that objective.


   Acknowledgements
 
Partially supported by grants CICYT-SAF 99/0102 and FIS 01/3135. A.R.-S. has a grant from the Ministerio de Sanidad y Consumo (99/3142).


   Notes
 
Time for primary review 9 days


   References
Top
 Abstract
 1. Structure and function...
 2. GJMIC in the...
 3. GJMIC during myocardial...
 4. Gap junction-mediated spread...
 5. GJ and ischemic...
 6. Conclusion and therapeutic...
 References
 

  1. Harris A.L. Emerging issues of connexin channels: biophysics fills the gap. Q. Rev. Biophys. (2001) 34:325–472.[ISI][Medline]
  2. Sosinsky G. Gap junctions. Molecular basis of cell communication in health and disease. Peracchia C., ed. (2000) vol. 49. San Diego: Academic Press. 1–22. Curr. Top. Membr.
  3. Brink P.R., Valiunas V., Christ G.J. Gap junctions. Molecular basis of cell communication in health and disease. Peracchia C., ed. (2000) vol. 49. San Diego: Academic Press. 43–60. Curr. Top. Membr.
  4. Bukauskas F.F., Peracchia C. Gap junctions. Molecular basis of cell communication in health and disease. Peracchia C., ed. (2000) vol. 49. San Diego: Academic Press. 207–221. Curr. Top. Membr.
  5. Veenstra R.D. Gap junctions. Molecular basis of cell communication in health and disease. Peracchia C., ed. (2000) vol. 49. San Diego: Academic Press. 95–129. Curr. Top. Membr.
  6. Skerrett I.M., Smith J.F., Nicholson B.J. Gap junctions. Molecular basis of cell communication in health and disease. Peracchia C., ed. (2000) vol. 49. San Diego: Academic Press. 249–269. Curr. Top. Membr.
  7. Moreno A.P., Chanson M., Anumonwo J., et al. Role of the carboxyl terminal of connexin43 in transjunctional fast voltage gating. Circ. Res. (2002) 90:450–457.[Abstract/Free Full Text]
  8. Bukauskas F.F., Bukauskiene A., Verselis V.K. Conductance and permeability of the residual state of connexin43 gap junction channels. J. Gen. Physiol. (2002) 119:171–185.[Abstract/Free Full Text]
  9. Bukauskas F.F., Peracchia C. Two distinct gating mechanisms in gap junction channels: CO2-sensitive and voltage-sensitive. Biophys. J. (1997) 72:2137–2142.[Abstract/Free Full Text]
  10. Bevans C.G., Kordel M., Rhee S.K., Harris A.L. Isoform composition of connexin channels determines selectivity among second messengers and uncharged molecules. J. Biol. Chem. (1998) 273:2808–2816.[Abstract/Free Full Text]
  11. Valiunas V., Beyer E.C., Brink P.R. Cardiac gap junction channels show quantitative differences in selectivity. Circ. Res. (2002) 91:104–111.[Abstract/Free Full Text]
  12. Veenstra R.D., Wang H.Z., Beyer E.C., Brink P.R. Selective dye and ionic permeability of gap junction channels formed by connexin45. Circ. Res. (1994) 75:483–490.[Abstract/Free Full Text]
  13. Beblo D.A., Veenstra R.D. Monovalent cation permeation through the connexin40 gap junction channel. Cs, Rb, K, Na, Li, TEA, TMA, TBA, and effects of anions Br, Cl, F, acetate, aspartate, glutamate, and NO3. J. Gen. Physiol. (1997) 109:509–522.[Abstract/Free Full Text]
  14. Veenstra R.D., Wang H.-Z., Beblo D.A., et al. Selectivity of connexin-specific gap junctions does not correlate with channel conductance. Circ. Res. (1995) 77:1156–1165.[Abstract/Free Full Text]
  15. Wang H.-Z., Veenstra R.D. Monovalent ion selectivity sequences of the rat connexin43 gap junction channel. J. Gen. Physiol. (1997) 109:491–507.[Abstract/Free Full Text]
  16. Dhein S. Gap junction channels in the cardiovascular system: pharmacological and physiological modulation. Trends Pharmacol. Sci. (1998) 19:229–241.[CrossRef][Medline]
  17. Kwak B.R., Jongsma H.J. Regulation of cardiac gap junction channel permeability and conductance by several phosphorylation conditions. Mol. Cell Biochem. (1996) 157:93–99.[CrossRef][ISI][Medline]
  18. Kwak B.R., van Veen T.A., Analbers L.J., Jongsma H.J. TPA increases conductance but decreases permeability in neonatal rat cardiomyocyte gap junction channels. Exp. Cell Res. (1995) 220:456–465.[CrossRef][ISI][Medline]
  19. Van Been T.A.B., van Rijen H.V.M., Opthof T. Cardiac gap junction channels: modulation of expression and channel properties. Cardiovasc. Res. (2001) 51:217–229.[Abstract/Free Full Text]
  20. Lampe P.D., Lau A.F. Regulation of gap junctions by phosphorylation of connexins. Arch. Biochem. Biophys. (2000) 384:205–215.[CrossRef][ISI][Medline]
  21. Dhein S., Polontchouk L., Salameh A., Haefliger J.-A. Pharmacological modulation and differential regulation of the cardiac gap junction proteins connexin 43 and connexin 40. Biol. Cell (2002) 94:409–422.[CrossRef][ISI][Medline]
  22. Jongsma H.J., van Rijen H.V.M., Kwak B.R., Chanson M. Gap junctions. Molecular basis of cell communication in health and disease. Peracchia C., ed. (2000) vol. 49. San Diego: Academic Press. 131–144. Curr. Top. Membr.
  23. Lampe P.D., Qiu Q., Meyer R.A., et al. Gap junction assembly: PTX-sensitive G proteins regulate the distribution of connexin43 within cells. Am. J. Physiol. (2001) 281:C1211–C1222.[ISI]
  24. Gaietta G., Deerinck T.J., Adams S.R., et al. Multicolor and electron microscopic imaging of connexin trafficking. Science (2002) 296:503–507.[Abstract/Free Full Text]
  25. Laing J.G., Beyer E.C. Gap junctions. Molecular basis of cell communication in health and disease. Peracchia C., ed. (2000) vol. 49. San Diego: Academic Press. 23–41. Curr. Top. Membr.
  26. Lin J.H.-C., Weigel H., Cotrina M.L., et al. Gap-junction-mediated propagation and amplification of cell injury. Nat. Neurosci. (1998) 1:494–500.[CrossRef][ISI][Medline]
  27. Krutovskikh V.A., Piccoli C., Yamasaki H. Gap junction intercellular communication propagates cell death in cancerous cells. Oncogene (2002) 21:1989–1999.[CrossRef][ISI][Medline]
  28. Kim D.Y., Kam Y., Koo S.K., Joe C.O. Gating connexin43 channels reconstituted in lipid vesicles by mitogen-activated protein kinase phosphorylation. J. Biol. Chem. (1999) 274:5581–5587.[Abstract/Free Full Text]
  29. Kanemitsu N.Y., Jiang W., Eckhart W. Cdc2-mediated phosphorylation of the gap junction protein, connexin43, during mitosis. Cell Growth Differ. (1998) 9:13–21.[Abstract]
  30. Lampe P.D., TenBrock E.M., Burt J.M., Kurata W.E., Johnson R.G., Lau A.F. Phosphorylation of connexin43 on serine368 by protein kinase C regulates gap junctional communication. J. Cell Biol. (2000) 149:1503–1512.[Abstract/Free Full Text]
  31. Trosko J.E. The role of stem cells and gap junctional intercellular communication in carcinogenesis. J. Biochem. Mol. Biol. (2003) 36:43–48.[ISI][Medline]
  32. Dhein S., Manicone N., Muller A., et al. A new synthetic antiarrhythmic peptide reduces dispersion of epicardial activation recovery interval and diminishes alterations of epicardial activation patterns induced by regional ischemia. A mapping heart. Naunyn Schmiedebergs Arch. Pharmacol. (1994) 350:174–184.[ISI][Medline]
  33. Muller A., Schaefer T., Linke W., et al. Actions of the antiarrhythmic peptide AAP10 on intercellular coupling. Naunyn Schmiedebergs Arch. Pharmacol. (1997) 356:76–82.[CrossRef][Medline]
  34. Weng S., Lauven M., Schaefer T., Polontchouk L., Grover R., Dhein S. Pharmacological modification of gap junction coupling by an antiarrhythmic peptide via protein kinase C activation. FASEB J. (2002) 16:1114–1116.[Abstract/Free Full Text]
  35. Xing D., Kjolbye A.L., Nielsen M.S., et al. ZP123 increases gap junctional conductance and prevents reentrant ventricular tachycardia during myocardial ischemia in open chest dogs. J. Cardiovasc. Electrophysiol. (2003) 14:510–520.[CrossRef][ISI][Medline]
  36. Schmidtmann M. Ueber die intracellulaere wasserstofftonenkonzentration unter physiologischen und einigen pahologischen bedingungen. Z. Gesammte Med. (1925) 45:714–742.[CrossRef]
  37. Revel J.P., Karnovsky M.J. Hexagonal array of subunits in intercellular junctions of the mouse heart and liver. J. Cell Biol. (1967) 33:C7–C12.[Free Full Text]
  38. Davis L.M., Kanter L., Beyer E.C., Saffitz J.E. Distinct gap junction protein phenotypes in cardiac tissues with disparate conduction properties. J. Am. Coll. Cardiol. (1994) 24:1124–1132.[Abstract]
  39. Davis L.M., Rodefeld M.E., Green K., Beyer E.C., Saffitz J.E. Gap junction protein phenotypes of the human heart and conduction system. J. Cardiovasc. Electrophysiol. (1995) 6:813–822.[ISI][Medline]
  40. Lin X., Crye M., Veenstra R.D. Regulation of connexin43 gap junctional conductance by ventricular action potentials. Circ. Res. (2003) 93:e63–e73.[Abstract/Free Full Text]
  41. Beblo D.A., Wang H.-Z., Beyer E.C., Westphale E.M., Veenstra R.D. Unique conductance, gating, and selective permeability properties of gap junction channels formed by connexin40. Circ. Res. (1995) 77:813–822.[Abstract/Free Full Text]
  42. Veenstra R.D., Wang H.-Z., Westphale E.M., Beyer E.C. Multiple connexins confer distinct regulatory and conductance properties of gap junctions in developing heart. Circ. Res. (1992) 71:1277–1283.[Abstract/Free Full Text]
  43. Severs N.J. The cardiac gap junction and intercalated disc. Int. J. Cardiol. (1990) 26:137–173.[CrossRef][ISI][Medline]
  44. Jongsma H.J., Wilders R. Gap junctions in cardiovascular disease. Circ. Res. (2000) 86:1193–1197.[Abstract/Free Full Text]
  45. Lo C.W. Role of gap junctions in cardiac conduction and development. Insights from the connexin knockout mice. Circ. Res. (2000) 87:346–348.[Free Full Text]
  46. Delmar M., Michaels D.C., Johnson T., Jalife J. Effects of increasing intercellular resistance on transverse and longitudinal propagation in sheep epicardial muscle. Circ. Res. (1987) 60:780–785.[Abstract/Free Full Text]
  47. Dhein S., Krusemann K., Schaefer T. Effects of the gap junction uncoupler palmitoleic acid on the activation and repolarization wavefronts in isolated rabbit hearts. Br. J. Pharmacol. (1999) 128:1375–1384.[CrossRef][ISI][Medline]
  48. Sperelakis N. An electric field mechanism for transmission of excitation between myocardial cells. Circ. Res. (2002) 91:985–987.[Free Full Text]
  49. McCallister L.P., Trapukdi S., Neely J.R. Morphometric observation on the effects of ischemia in the isolated perfused rat heart. J. Mol. Cell Cardiol. (1979) 11:619–630.[CrossRef][ISI][Medline]
  50. De Mello W.C. Cell-to-cell coupling assayed by means of electrical measurements. Experientia (1987) 43:1075–1079.[CrossRef][ISI][Medline]
  51. De Mello W.C. The influence of pH on the healing-over of mammalian cardiac muscle. J. Physiol. (1983) 339:299–307.[Abstract/Free Full Text]
  52. Delage B., Deleze J. Gap junctions. Molecular basis of cell communication in health and disease. Peracchia C., ed. (2000) vol. 49. San Diego: Academic Press. 189–206. Curr. Top. Membr.
  53. Kleber A.G., Riegger C.B., Janse M.J. Electrical uncoupling and increase of extracellular resistance after induction of ische