Are there gap junctions in cardiac muscle

Ac Histograms of interdetector conduction times indicate coexistence of fast cytoplasmic and delayed transjunctional activation. B Multiple cell wide strands: panels as in A. In the case of strands several cells wide, differences between cytoplasmic and transjunctional conduction times are substantially reduced indicating nearly continuous conduction along the preparation.

Redrawn with modifications from Ref. Characteristics of impulse propagation on the subcellular scale. A Cellular architecture of the optically mapped region of a chain of single cardiomyocytes during impulse propagation from left to right end-to-end abutted cardiomyocytes shown in two shades of gray. Squares indicate the positions of individual photodetectors. B Simultaneously recorded action potential upstrokes indicate a local delay between detectors 5 and 7, i.

Redrawn with permission from Ref. Computer simulation of microscopic impulse propagation. A Action potential upstrokes recorded during impulse propagation between two adjacent cells at the sites indicated in the insert. Under conditions of normal gap junctional coupling, the transjunctional conduction delay is roughly equal to the myoplasmic conduction times of the cells.

B During a tenfold reduction of gap junctional coupling, myoplasmic conduction times are abbreviated whereas transjunctional conduction times are substantially increased.

Whereas, as noted above, propagation in uniform multicellular tissue under normal conditions is continuous, this changes during progressive uncoupling. Also at the cellular level, impulse propagation under these conditions is highly discontinuous as illustrated in Fig.

Moreover, when mapping the clustered action potential upstrokes to the cellular structure of the preparation Fig. Thus, irrespective of scaling, meandering activation seems to be a basic principle governing impulse propagation during reduced gap junctional coupling in cardiac tissue. Effects of critical gap junctional uncoupling on impulse propagation characteristics at the cellular level.

Bc Schematic representation of the path of activation and its dependence on the cellular tissue architecture. Regions showing quasi-simultaneous activation are numbered according to the cluster numbers indicated in Bb. The hatched region indicates a completely uncoupled cell showing no electrical activation. Characteristics of conduction during severe gap junctional uncoupling in multiple cell wide strands.

B During severe gap junctional uncoupling, activation invades the preparation in a meandering fashion, which is due to the presence of electrically insulated, i. C Whereas activation is continuous and fast during normal gap junctional coupling The reason for this behavior has been analyzed in detail in computer simulation studies by Shaw and Rudy [12,24] , which compared the characteristics of conduction slowing during a reduction of either excitability or gap junctional coupling.

By this definition, conduction fails when the safety factor drops below 1 and becomes increasingly stable as it rises above 1. In contrast to these effects, a reduction of intercellular coupling showed marked differences Fig. First, whereas progressive uncoupling also induced a monotonic decrease of conduction velocity, slowest conduction velocities reached before block 0. Moreover, both maximal upstroke velocities and the safety factor for conduction showed a biphasic behavior with a substantial initial increase during progressive uncoupling.

This transient increase of maximal upstroke velocities and of the safety factor of conduction above control values is due to the fact that, with decreasing gap junctional conductance, the sodium inward current is increasingly confined to individual cells because less current is lost downstream.

On the other hand, the decrease in the safety factor for conduction and maximal upstroke velocity at very low levels of gap junctional coupling is caused by the highly reduced axial current flow downstream, which causes long subthreshold charging of the cells ahead and, concomitantly, a progressive inactivation of sodium channels. Computer simulation of the effects of a gradual reduction of gap junctional coupling on the characteristics of impulse propagation.

A Dependence of conduction velocity and of the safety factor for conduction on the degree of intercellular coupling. B Dependence of maximal upstroke velocity and maximal sodium inward current on the degree of intercellular coupling.

Computer simulation of the effects of a gradual reduction of membrane excitability on the characteristics of impulse propagation. A Dependence of conduction velocity and of the safety factor for conduction on membrane excitability.

B Dependence of maximal upstroke velocity and maximal sodium inward current on membrane excitability. In conclusion, whereas impulse propagation under physiologic conditions along single cell chains of cardiomyocytes is saltatory due to the recurrent increases in axial resistance at the sites of gap junctional coupling, this feature is lost in intact multicellular tissue due to lateral gap junctional coupling which serves to average local small differences in activation times of individual cardiomyocytes at the excitation wavefront.

In multicellular tissue, saltatory conduction only reappears under conditions of critical gap junctional uncoupling. There it leads to a functional unmasking of the cellular structure and induces ultra-slow and meandering conduction, which is well known to be a key ingredient in arrhythmogenesis. In both experiments and computer simulations, partial gap junctional uncoupling was shown to result in conduction velocities, which are over an order of magnitude slower than those obtained during a maximal reduction of excitability.

The only feature of the characteristics of impulse propagation during severe uncoupling in computer simulation studies not reproduced routinely by experiments so far for exception, cf. This is most probably due to the lack of specific uncoupling agents available, because increases in maximal upstroke velocities accompanying a reduction in conduction velocity have been found in mice with connexin43 null mutations [26]. Such tissue structures induce propagation delays in anterograde direction due to the presence of a source-to-sink mismatch.

If the mismatch is large enough, unidirectional conduction blocks ensue. Computer simulations of the characteristics of propagation across such expansions suggested that, contrary to intuition, partial gap junctional uncoupling might improve conduction and remove unidirectional conduction blocks [27,28].

This prediction has been investigated with patterned growth cell cultures as illustrated in Fig. The preparations consisted of narrow strands of cardiomyocytes, which merged with a large rectangular cell sheet Fig.

The widths of the strands were chosen such that unidirectional conduction block occurred under control conditions Fig. Initially, complete gap junctional uncoupling of the strand and the expansion was induced by local superfusion with palmitoleic acid Fig. Thereafter, washout was started which resulted in a gradual increase of gap junctional coupling during which successful bi-directional conduction could be observed within a limited time window Fig. After reaching normal levels of gap junctional coupling, unidirectional conduction blocks were reestablished Fig.

This illustrates that partial gap junctional uncoupling has the potential to remove unidirectional conduction blocks. This counter-intuitive behavior can be explained by differences in the dimensionality of the effect of gap junctional uncoupling on the source strand and the sink expansion.

Partial uncoupling of the essentially linear source will reduce its size in a linear proportional manner whereas equal uncoupling of the two-dimensional sink will reduce the size thereof following a square function. This overproportional reduction of the sink improves the source-to-sink mismatch up to the point of successful anterograde conduction.

As can be expected for a gradual change in the balance between the source and the sink, overall conduction velocity in anterograde direction across a tissue expansion changes in a biphasic manner during progressive re-coupling. This is shown in Fig. How can this be explained? At the beginning of the establishment of successful anterograde conduction, conduction velocities are initially in the range of those observed in linear cell strands cf.

This suggests that conduction at these very low levels of gap junctional conductance is primarily determined by the degree of cell-to-cell coupling and not by the source-to-sink mismatch represented by the expansion. Accordingly, increases of gap junctional conductance led initially to an increase in conduction velocity. At the same time, however, the parallel increase in the size of the sink became more important and, ultimately, resulted in the paradoxical situation that overall conduction velocity decreased even though gap junctional conductance increased.

Therefore, in contrast to the situation in uniform cell structures where a decrease in gap junctional coupling is invariably accompanied by a decrease of conduction velocity, overall conduction velocity across sites where planar wavefronts change to curved wavefronts expansion, isthmus , initially show an increase which is followed by a decrease only during severe uncoupling.

Development of macroscopic conduction velocity across a tissue expansion during a gradual increase of gap junctional conductance same experimental protocol as in Fig. Induction of successful bi-directional conduction across a tissue expansion by partial gap junctional uncoupling. A Schematic drawing of the experimental layout.

A tissue expansion was locally superfused with the gap junctional uncoupler palmitoleic acid and impulse propagation characteristics across the expansion were assessed optically by detectors black rings along the central axis during anterograde left and retrograde right conduction. B Action potential upstrokes recorded under control conditions indicate decremental conduction during anterograde and normal conduction during retrograde stimulation, i.

C Complete gap junctional uncoupling results in bi-directional conduction blocks at the boundaries of the superfusion. D During gradual re-coupling, successful bi-directional conduction is established.

E At the end of the washout period, normal gap junctional conductance is reestablished as indicated by fast retrograde conduction and anterograde conduction block. Thus, whereas gap junctional uncoupling in general leads to an impairment of conduction and thereby contributes significantly to the generation of arrhythmias, there is another side to this coin for the case of discontinuous tissue architectures. Accordingly, while improvement of gap junctional coupling in structurally diseased heart might be expected to act in an anti-arrhythmogenic manner by reducing the incidence of slowly conducting pathways, it might at the same time provoke arrhythmias by unmasking potential regions of unidirectional conduction blocks.

One of the prominent biophysical features of gap junctions is their time- and voltage-dependent inactivation. In another recent study using transfected neuroblastoma cells, inactivation kinetics of connexin43 were studied by imposing an action potential clamp instead of a rectangular voltage pulses on one of the cells [34]. In addition to the effects of gap junctional gating on depolarizing current flow in the orthodromic direction, it would be interesting to know whether gating possibly affects the trailing part of activation, i.

While it is tempting to speculate that partial inactivation might last into the repolarization phase, thereby channeling depolarizing current from the activation wavefront in the orthodromic direction, this mechanism is unlikely to exist because gap junctional conductance will rapidly recover during repolarization due to the reversal of polarity [35] and the decrease in size of the transjunctional voltage [34]. Thus, whereas gating possibly influences the degree of ultra-slow conduction to some extent during severe uncoupling, the role thereof during normal propagation is likely to be insignificant.

However, because gating is dependent on a variety of factors such as the expression system used, transjunctional voltages present, and the type of gap junctions present isoforms, connexon composition homomeric, heteromeric and connexon coupling homotypic, heterotypic definitive answers to this question will require further studies, which take into account transjunctional voltage differences as they occur across a given junction during propagated activity.

In the heart, approximately half of the cells consist of non-myocardial cells, among which fibroblasts constitute the largest fraction [36]. This number can be expected to increase as a result of cardiac diseases leading to fibrosis. Whereas the formation of excessive collagen sheets, which act as electrical insulators, has been recognized for a long time as being a cause for discontinuous conduction and the occurrence of arrhythmias [37] , the question of whether the cellular constituents of fibrotic tissue, i.

It has been known for several decades that individual fibroblasts of cardiac origin can establish gap junctional communication with adjacent cardiomyocytes in culture [38,39].

In this context, it was shown that impulse propagation in monolayer cardiomyocyte cultures can be modified by grafting a layer of fibroblasts transfected with the voltage gated potassium channel Kv1. This co-culture resulted in conduction blocks in the cardiomyocyte monolayer, which were reversed upon application of specific blockers of the potassium channel. Moreover, it was shown in cell culture that fibroblasts adjacent to cardiomyocytes induce a decrease in maximal upstroke velocity [41] or a local slowing of conduction [42].

Whereas these findings illustrate that fibroblasts in intimate contact with cardiomyocytes can influence the electrophysiological behavior of the latter via gap junctions, the question arises whether such interactions might also occur over longer distances, i.

This issue was recently investigated with a cell culture model where patterned growth strands of cardiomyocytes were interrupted over defined distances by fibroblasts of cardiac origin [43].

The results of one of these experiments are shown in Fig. As indicated by the optically recorded transmembrane voltage signals during propagated activity in Fig. As shown by immunocytochemistry, this electrotonic interaction was based on the presence of both connexin43 and connexin As indicated by the plot of activation times along the preparation Fig.

Whereas it is tempting to speculate that such extremely slow conduction might be instrumental in the generation of arrhythmias in fibrotic hearts, studies with cell cultures have to be interpreted cautiously in regard to extrapolation of the results to intact tissue because there is as yet no firm proof of gap junctional coupling between cardiomyocytes and fibroblasts in-vivo. In contrast, a thorough investigation of this question found no evidence for robust gap junctional coupling in healthy intact tissue [44].

This raises the question whether fibroblasts in culture might undergo a phenotype switch to so-called myofibroblasts, which enables them to form gap junctions with cardiomyocytes. In intact hearts, it was shown that fibroblasts convert into myofibroblast after a local loss of cardiomyocytes [46]. Most interestingly, this conversion has been described to be accompanied by the expression of connexin43 in the case of breast cancer stroma cells [47] and myofibroblasts derived from corneal fibroblasts [48].

If such a conversion of fibroblasts into connexin expressing myofibroblasts should also occur in the heart under pathophysiological conditions such as myocardial infarction [36] , this would raise the interesting hypothesis that the ensuing coupling of non-excitable cells to cardiomyocytes might lead to very slow conduction and, thus, might constitute a possibly important new arrhythmogenic mechanism. Impulse transmission along cardiac fibroblasts. B Staining with anti-myomesin antibodies shows the boundaries of the cardiomyocyte strands.

C Optically recorded action potential upstrokes along the preparation show monotonically rising signals in the region of cardiomyocytes black and biphasic upstrokes in the region of fibroblasts gray. Signals from the distal cardiomyocyte strand show pronounced subthreshold depolarization. In conclusion, the observation that cardiomyocytes readily form functional gap junctions with heterogeneous cells and that this coupling supports the spread of excitation over extended distances may have implications beyond electrical interactions with fibroblasts as presented above.

In particular, given the recent interest in using stem cells as a therapeutic approach for the diseased heart, the promiscuous gap junctional coupling is a prerequisite both for permitting orderly excitation sequences in the regions of grafted cells and for the intercellular exchange of signaling molecules. Whereas all of the above evidence stresses the importance of gap junctional coupling for impulse propagation under both physiologic and pathophysiologic conditions, the observation that the main ion channels underlying fast conduction, i.

If one were to design a cardiomyocyte, one would probably not plan to insert sodium channels at the intercalated disc because, among other reasons, they would face a highly restricted extracellular space which could be expected to be subject to large fluctuations in ion concentration and, consequently, in adverse changes in electrochemical driving forces.

On the other hand and as formulated many years ago by Sperelakis and Mann [54] , the fact that space is restricted at the intercalated disc could also act in favor of impulse propagation.

These authors postulated an electric field mechanism of impulse transfer, which is based on the idea that activation of sodium channels located at the intercalated disc results in a negative shift of the cleft potential between a given activated cardiomyocyte and a neighboring quiescent cardiomyocyte. Sperelakis and McConnell [51] further suggested that this mechanism might be supported by rapid potassium accumulation in the cleft during activation of the pre-junctional cell which would induce a depolarization of the post-junctional membrane cell to threshold.

Obviously, this electric field mechanism is critically dependent on the radial shunt resistance of the intercellular cleft which has i to assume a value high enough to permit the build-up of a local extracellular negativity in the cleft region and ii which, at same time, must permit establishment of a local circuit current large enough as to depolarize the quiescent cell.

Recently, the interplay between the effects of sodium channel clustering, cleft potentials and gap junctional coupling on impulse propagation have been further investigated by using a linear strand of cardiomyocytes represented by the Luo—Rudy model [52]. The effects of different combinations of cleft widths and degrees of sodium channel partitioning on conduction velocities achieved during progressive uncoupling are shown in Fig.

Under the assumption of a cleft width of 35 nm and sodium channels being present exclusively at the intercalated disc, these simulations showed that conduction velocities were substantially slower than those obtained with a non-cleft model during normal gap junctional coupling. This was explained by the occurrence of large negative cleft-potentials during activation, which induced a rapid and largely overshooting response of pre- and post-junctional membranes and, therefore, resulted in an attenuation of the sodium inward current and a concomitant slowing of conduction.

Whereas this finding of conduction in the virtual absence of gap junctional coupling points in the same direction as earlier findings by Sperelakis and colleagues, Fig. Whereas these studies show the possibility of conduction in the absence of gap junctional coupling, it is not clear whether these findings are relevant for intact tissue. Whereas clustering of sodium channels at the intercalated discs is undisputed for both intact cardiac tissue and cell cultures, there are no experimental data available regarding actual radial cleft resistances in cardiac tissue.

Because a direct determination of the radial cleft resistance is out of reach of present experimental methods, an exact morphometric assessment of the three-dimensional architecture of the cleft between two adjoining cardiomyocytes might indirectly permit one to obtain an approximate estimate thereof.

However, even if such studies should reinforce the computer simulation studies, there remains a number of open questions which are difficult to reconcile with the concept that electric field mechanisms alone are significantly involved in cardiac impulse conduction: i in the absence of substantial gap junctional coupling, space constants should be much shorter than what was generally reported in the past for a critical discussion of this issue, cf. Dependence of conduction velocity on gap junctional coupling for 3 different models of cleft configuration and partitioning of sodium channels.

Solid line: model with no cleft-effects but clustering of all sodium channels at intercalated disc. Dashed line: model with nm wide cleft and even partitioning of the sodium channels between the intercalated disc and the surface sarcolemma. Punctate line: model with a nm wide cleft and clustering of all sodium channels at intercalated disc.

Nevertheless, the clustering of sodium channels at the intercalated disc together with the presence of Na—K-ATPases at the same location [56] remains a highly intriguing and interesting fact. It indicates that this part of the sarcolemma, even though facing a highly restricted extracellular space, might exhibit a specific function.

However, answers as to the exact nature and relevance of this function will have to await further studies. Gap junctional coupling plays a crucial role for impulse propagation in cardiac tissue. It is well established that a reduction of gap junctional conductance as occurring, e. At the same time, however, partial gap junctional uncoupling has the capacity to remove unidirectional conduction blocks and therefore acts, at least over a certain range of partial uncoupling, in an anti-arrhythmogenic fashion.

This suggests that electrical uncoupling is not inevitably followed by arrhythmias. Rather, it ultimately depends on the balance between pro- and antiarrhythmic effects of uncoupling whether a certain reduction of axial current flow in the setting of non-uniform tissue structures can give rise to reentrant excitation.

A second possibly pro-arrhythmic mechanism caused by gap junctions in the diseased heart might be related to heterogeneous cell coupling between cardiomyocytes and fibroblasts. Whereas it has been shown in cell culture that fibroblasts are able to relay electrical activation with substantial delays over appreciable distances, it remains to be shown whether gap junctional coupling between heterogeneous cell populations is indeed present in the diseased heart. Finally, since the observation of a clustering of sodium channels at the intercalated disc, the hypothesis that there might exist gap-junction independent mechanisms of impulse propagation in cardiac tissue has received renewed attention.

Because current transfer occurs only at gap junctions, the spatial distribution and biophysical properties of gap junction channels are important determinants of the conduction properties of cardiac muscle. Gap junction channels are composed of members of a multigene family of proteins called connexins. As a general rule, individual cells express multiple connexins, which creates the potential for considerable functional diversity in gap junction channels.

Although gap junction channels are relatively nonselective in their permeability to ions and small molecules, cardiac myocytes actively adjust their level of coupling by multiple mechanisms including changes in connexin expression, regulation of connexin trafficking and turnover, and modulation of channel properties. The wave of contraction that allows the heart to work as a unit, called a functional syncytium, begins with the pacemaker cells.

This group of cells is self-excitable and able to depolarize to threshold and fire action potentials on their own, a feature called autorhythmicity ; they do this at set intervals which determine heart rate. Privacy Policy. Skip to main content. Module Muscle Tissue. Search for:. Licenses and Attributions. CC licensed content, Shared previously.