Clarifying the Complexities of Connexins and Conduction
See related article, pages 1216–1224
In the late 19th century microscopy and early physiological experimentation gave rise to the apparent paradox that although cardiac muscle is made up of individual cells, it behaves as a single continuous functional unit. The explanation had to be that a component of the individual cells was responsible for coupling them together and coordinating their function to provide one continuous electromechanical unit. In the 1960s high resolution microscopy identified the likely membrane specialization responsible for this, and the functional properties of what later became known as the “gap junction” started to be determined by experimentation in the 1970s.
Gap-junctional membrane offers relatively low resistance to current flow, several orders of magnitude lower than ordinary cell membrane. But in passing through the cytoplasmic pathway from cell to cell in whole tissue, gap junctions remain relatively resistive discontinuities to the passage of ions and electrical charge through this cytoplasmic pathway, presenting a resistance across the gap junction (of nanometre width) approximately equivalent to that of the column of cytoplasm of an entire cell length (≈100 μm).
Although electrophysiological, molecular and genetic techniques have been used to provide a very substantial body of knowledge of gap-junctional structure and function the precise mechanism by which the action potential is propagated from cell to cell, and the precise role of the gap junction remains unclear. It is thought to determine how much depolarizing current can pass from a depolarized cell to its neighbor in the process of impulse propagation, thus providing continuity across the cell-cell interface of the local electromagnetic flux and field which triggers the membrane changes and therefore action potential generation through adjacent cells. The 3-dimensional topology of gap junctions, itself determined by the morphology and packing geometry of the constituent cells, is thus a principal determinant of the resistive and, therefore, conduction properties of myocardium.
With respect to the morphological determinants of myocardial conduction, the 3-dimensional distribution of gap junctions represents a basic level of organization. The article by Beauchamp et al1 in this issue provides clarification of further levels of the complexity and dynamics of the functional morphology of conduction, and addresses many of the important questions relating to the role of the gap-junctional proteins (connexins) in atrial conduction. This study shows that interaction exists between the two dominant atrial connexins (connxin43 [Cx43] and connexin40 [Cx40]) when coexpressed, and that this interaction is not simple and is modulated by changes in the relative distributions of the connexins between junctional (and therefore available for cell-cell coupling) and nonjunctional compartments of the cell.
Specifically, what Beauchamp et al1 have demonstrated is that the response to Cx43 reduction and elimination is that although total cellular Cx40 protein levels remain unchanged (or even increased) there is a shift of Cx40 to being nonjunctionally located with a consequent reduction in gap-junctional Cx40 (−50%) and reduction in CV (−44%). By contrast, with Cx40 reduction and elimination the unchanged total Cx43 content shifts to being more junctionally-located Cx43 (+160%), but as discussed by the authors, the consequent increase in conduction velocity (+50%) can only be adequately explained by postulating that when Cx40 and Cx43 are coexpressed, the resulting intercellular conductance is less than either connexin alone, and that the interaction between connexins 40 and 43 when coexpressed is, therefore, complex and apparently inhibitory. The result of complete elimination of Cx40 is a high conduction velocity closely approximating that in the ventricle, as would be expected on the basis of Cx43 being the sole dominant connexin in both ventricular and Cx40-depleted atrial myocardium, and the authors conclude that dominance of Cx40 decreases and dominance of Cx43 increases conduction velocity in patterned cultures of neonatal atrial myocytes (Table).
Based on limited experimental information from intact myocardium, it has long been reasonable to hypothesize that changes in gap-junctional coupling are central to arrhythmogenesis for 2 reasons: 1) as the cause of the slow conduction characteristic of re-entrant arrhythmogenesis; and 2) by limiting the passage of current between adjacent cells thereby unmasking their inherent differences in action potential characteristics and increasing heterogeneity, thereby promoting arrhythmogenesis by mechanisms involving action potential abnormalities (triggered activity and automaticity) as well as reentrant conduction. Mathematical and computer modeling supported these hypotheses,2,3 but it was not until the last decade or so that direct experimental data have shed light on the relationship between gap-junctional topology and function, conduction and arrhythmogenesis,4–6 adding to the body of evidence in support of these hypotheses, and dispelling some of the myths associated with this cell membrane specialization that has for so long been identified but for which it has until recently proven difficult to get experimental data of direct relevance to in vivo conduction.
The study by Beauchamp et al1 very conclusively adds to the body of evidence against a previously held concept of there being overwhelming redundancy of gap-junctional coupling and that uncoupling of the order of >90% may be necessary to affect conduction velocity. This concept gained some popularity among researchers in the field in the 1990s and resulted in a tendency to dismiss studies reporting relatively subtle changes in connexin levels and distribution in diseased and arrhythmogenic myocardium,7,8 as being biologically irrelevant. What we now know from this and other well conducted studies1,4 is that there is little or no apparent redundancy or threshold level of uncoupling below which there is no detectable effect on conduction, but more that there is a continuous coupling-dependent effect on conduction, and that gap-junctional uncoupling to a lesser (and even normally-variant5) degree may modify conduction particularly in stressed systems in which cell function may otherwise be modified, such as under pathological conditions.
What the study by Beauchamp et al1 has also elucidated is a potential mechanism by which atrial conduction may be modulated by the relative expression of the 2 dominant connexins, and that variation in gap-junctional Cx40 may have an inverse relationship with conduction velocity. As stated by the authors, this would be consistent with observations made on the relationship between conduction velocity and natural variation in gap-junctional connexin levels in human atrium.5 The question then is what, if any, role changes in connexin quantity and distribution play in atrial fibrillation? It would appear that although connexin levels do not correlate with fibrillation wave front propogation velocity, they correlate with the complexity of activation (the number of wave fronts per unit area) and it would seem likely that in conditions of incomplete excitability resulting from partially excitable gaps, such as occurs in fibrillation in remodelled atria, connexin levels become an important determinant of the patterns of conduction and wave break.6
Despite the limitations imposed by the degree to which genetic engineering of protein expression and neonatal cell culture are representative of intact adult myocardium in vivo, the patterning of the cell growth and the spatial resolution of the optical mapping system used by this group has in the past provided many valuable insights into the relationship between connexins and conduction,9,10 and continues to provide a powerful tool for understanding the biology of gap junctions. Genetic modification remains useful for investigating the functional effects of modulating protein expression, and provided there are no other confounding changes consequent on manipulating levels of the index protein, data from such studies helps dissect out the basic role of the protein of interest and its interactions. The finding that Cx40 knockout alters Cx43 expression is of great interest but what is not known is whether these relationships of expression and distribution occur naturally in myocardium in vivo, or are simply a feature of the specific genetically-engineered model. In terms of clinical relevance, the authors cite our article on connexin levels and conduction as measured in the intact human atrium as being remarkably consistent,5 and a recent study has implicated a mutation in the Cx40 gene in human atrial fibrillation.11
The human atrium has a complex structure which at the macroscopic level varies between different regions such as trabecular versus smooth-walled regions, and at functional specializations such as the crista terminalis, Bachman’s bundle and the septo-pulmonary bundle of Papez,12 (all of which have distinct conduction properties), and which at the microscopic level has a three dimensional topology of gap-junctional discontinuities in the cytoplasmic pathway that is distinct and variable between regions. In contrast to the neonate, gap junctions are predominantly confined to discretely distributed intercalated disks in adult myocardium,13 and although the cell cultures used in the study by Beauchamp et al1 are devoid of any of these macroscopic or microscopic complexities of intact myocardium in vivo, it is the relative simplicity of the myocyte preparation used that is its strength in providing clarity and the ability to dissect out yet another individual component of the complex picture.
The picture that is emerging is of regulatory control of cell-cell coupling by which subtle changes in the relative expression or the degree of membrane incorporation of connexins will modulate coupling, and in conjunction with the known metabolic modulators of cell-cell coupling that include pH, Ca2+ concentration, oxygenation, and phosphorylation status, provide a number of modulating mechanisms that act in concert and over different time scales, and are further modified and disrupted by remodeling of myocardium in disease and arrhythmogenesis.
Sources of Funding
N.P. is funded by St. Mary’s Hospital Trust, Imperial College London, and the British Heart Foundation.
The opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association.
Beauchamp P, Yamada KA, Baertschi AJ, Green K, Kanter EM, Saffitz JE, Kleber AG. Relative contributions of connexins 40 and 43 to atrial impulse propagation in synthetic strands of neonatal and fetal murine cardiomyocytes. Circ Res. 2006; 99: 1216–1224.
Peters NS, Coromilas J, Severs NJ, Wit AL. Disturbed connexin43 gap junction distribution correlates with the location of reentrant circuits in the epicardial border zone of healing canine infarcts that cause ventricular tachycardia. Circulation. 1997; 95: 988–996.
Peters NS, Green CR, Poole-Wilson PA, Severs NJ. Reduced content of connexin43 gap junctions in ventricular myocardium from hypertrophied and ischemic human hearts. Circulation. 1993; 88: 864–875.
Thomas SP, Kucera JP, Bircher-Lehmann L, Rudy Y, Saffitz JE, Kleber AG. Impulse propagation in synthetic strands of neonatal cardiac myocytes with genetically reduced levels of connexin43. Circ Res. 2003; 92: 1209–1216.
Zhuang J, Yamada KA, Saffitz JE, Kleber AG. Pulsatile stretch remodels cell-to-cell communication in cultured myocytes. Circ Res. 2000; 87: 316–322.
Markides V, Schilling RJ, Ho SY, Chow AW, Davies DW, Peters NS. Characterization of left atrial activation in the intact human heart. Circulation. 2003; 107: 733–739.
Peters NS, Severs NJ, Rothery SM, Lincoln C, Yacoub MH, Green CR. Spatiotemporal relation between gap junctions and fascia adherens junctions during postnatal development of human ventricular myocardium. Circulation. 1994; 90: 713–725.