Edward Carmeliet: his contributions and scientific legacy

The death of Edward Carmeliet has robbed us of one of the founders of modern cardiac electrophysiology. Here we summarize his contributions and, equally importantly, consider the current situation of the fields which he helped to develop. It would, however, be remiss not to point out that from 1980 to 1987, he was a member of the Editorial Board of The Journal of Physiology, one of the many roles in which he contributed to helping others develop their science. The Journal has also been an important forum for his work and was his publication of choice for the full reports of his mechanistic studies of cardiac electrophysiology. The dazzling variety of topics which he contributed to is also noteworthy and ranges from the basic properties of Ca2+ and K+ currents, the control of action potential duration, excitation–contraction coupling, to translational aspects such as the effects of ischaemia, and antiarrhythmic therapy. Such breadth was perhaps more common in days gone by than today. Edward Carmeliet studied Cl− movements while working in Berne in Silvio Weidmann's laboratory. His work was published as a single-author paper (Carmeliet, 1961) in which he simply thanked Weidmann for ‘valuable help and criticism’. One wonders whether today's culture might have resulted in a longer author list. He found that Cl− removal had no effect on either the maximum diastolic potential or the resting potential, a result which differed markedly from findings which had been recently published in skeletal muscle (Hodgkin & Horowicz, 1959). Carmeliet returned to Cl− currents 30 years later (Sipido et al. 1993) showing the existence of a Ca2+-activated Cl− current in the heart in rabbit Purkinje cells. This current activates and deactivates quickly during release of calcium from the sarcoplasmic reticulum, and adds to the early repolarization of the action potential. When the sarcoplasmic reticulum is emptied with caffeine, an additional slow component is observed, consistent with a fast activation by local calcium followed by activation by the bulk cytosolic calcium (Papp et al. 1995). While the molecular basis of the Ca2+-activated Cl− current is now known (for recent review, see Varró et al. 2021), there is still a lack of clarity about its function. It is present in the ventricle of sheep and pig, but absent in rat and mouse where fast early repolarization results from activation of K+ currents. It is unclear what the situation is in the human and pharmacological evidence suggests little role (Verkerk et al. 2003). In species that do possess this current, the contribution to the fast phase of repolarization may increase the driving force for Ca2+ entry as has been shown for K+ currents (Sah et al. 2002). Edward Carmeliet made major contributions to our understanding of calcium signalling in the heart. He demonstrated that Ca2+ release from the SR inactivated the L-type Ca2+ current and suggested that recovery from this Ca2+-dependent inhibition might underlie abnormal early afterdepolarizations (EADs) (Sipido et al. 1995a). This work led to a comprehensive study of the Ca2+-dependent membrane currents underlying the so-called transient inward current, Iti. Under calcium-overload conditions, this current occurs at diastolic potentials resulting in a delayed afterdepolarization (Lederer & Tsien, 1976), responsible for many cardiac arrhythmias. At the time there were several calcium-dependent candidates for this current: Na+/Ca2+ exchange (NCX), the non-specific cation current, and Ca2+-activated Cl− current. Carmeliet and colleagues demonstrated that NCX was the major contributor (Sipido et al. 1995b). It is still a puzzle as to why the non-specific cation current, which was one of the first single channels to be identified with the then novel patch clamp technique (Colquhoun et al. 1981), does not contribute more to cardiac electrophysiology. Interestingly, however, there was a significant contribution from calcium-dependent inactivation of the L-type Ca2+ current confirming its potential role in early afterdepolarizations. Even today, it is unclear exactly what the mechanisms are that underlie EADs. To what extent are they, as was originally suggested, a product of reactivation of the L-type Ca2+ current and, if so, is this due to recovery from voltage- as opposed to calcium-dependent inactivation? Alternatively, does the EAD (like the delayed afterdepolarization; DAD) result from Ca2+ release from the SR activating inward membrane currents? In the presence of high sympathetic drive, one of the major triggers for arrhythmias, probably both mechanisms contribute to EADs, which then are also seen together with DADs (Volders et al. 2000). Carmeliet also made important contributions to the concept of microdomains near the membrane where ion concentrations differ from those in the bulk cytosol. This was triggered by his observations on activation of a Na+-dependent K+ current, and of gradients created by the Na+/K+ pump (Carmeliet, 1992a). This became a controversial area in the 1990s when it was proposed that local Na+ increase was sufficient for Ca2+ influx through the Na+/Ca2+ exchanger to trigger release of calcium from the sarcoplasmic reticulum (Leblanc & Hume, 1990). The work generated much discussion on appropriate experimental conditions and Carmeliet encouraged further studies (Sipido et al. 1995c). Lack of proper tools has hampered a resolution but, with the availability of new methods such as membrane-targeted probes (Shang et al. 2014), this may change. He also provided evidence for microdomains of calcium concentration by showing that the Ca2+-activated Cl− current decayed much more quickly than the measured change of bulk calcium concentration ([Ca2+]i), an effect attributed to local gradients of [Ca2+]i (see also Papp et al. 1995; Trafford et al. 1995). In 1955, as a medical student in the laboratory of Physiology, Edward Carmeliet published his first single author paper on ‘Influence of rhythm on the duration of the action potential’ (Carmeliet, 1955). This interest would remain throughout his life, with an authoritative book that brings together insights in the modulation of voltage-dependent channels, the influence of changes in calcium handling and the autonomic modulation of ion channels and calcium (Carmeliet & Vereecke, 2002). Changes in rhythm are the consequence of changes in autonomic drive, acting on the sinus node. Carmeliet made important contributions to our understanding of pacemaker mechanisms and the properties of the acetylcholine-sensitive K+ channel (Callewaert et al. 1984; Carmeliet & Mubagwa, 1986). This was a foundation for later work, which demonstrated the importance of this current in atrial fibrillation (Dobrev et al. 2001). Here downregulation of the channel reduces muscarinic regulation, further contributing to the pathology. Carmeliet also studied the α-adrenergic modulation of the action potential through noradrenaline, fuelled by an interest in ischaemia and the associated catecholamine release. The effects were different in atrium and ventricle and between species, related to differences in receptor types and distribution. This complexity was unravelled and clearly presented by Carmeliet in his comprehensive review of ionic currents during ischaemia (Carmeliet, 1999). Yet, still today, the α-adrenergic modulation in human ventricular myocytes is not as well understood as β-adrenergic modulation. Carmeliet started his studies of K+ currents using isotope flux measurements in Purkinje fibres and in embryonic chick hearts (Carmeliet et al. 1976). He next used the two-microelectrode voltage clamp of thin Purkinje fibres and direct current measurements (Coraboeuf & Carmeliet, 1982). Work in these multicellular tissues was hampered by changes of extracellular ion concentrations in intercellular spaces. Later experiments in isolated myocytes removed some of the complexity, and with the patch clamp technique a new era of direct K+ current recordings started (Carmeliet et al. 1987). Carmeliet had a strong interest in K+ channels and studied many members of this large family. His studies were of direct relevance to drug development. K+ currents are a prime target to modulate action potential duration, and the delayed rectifier current a strong candidate (Carmeliet, 1992b). His work on modulation of K+ and Na+ currents was at the heart of his involvement with a translational and multi-stakeholder task force on anti-arrhythmic drug development, which first met in Taormina, Sicily. After this first meeting, the group published a viewpoint, the Sicilian Gambit (Task Force of the Working Group on Arrhythmias of the European Society of Cardiology, 1991), proposing a new approach to anti-arrhythmic drugs after the devastating results of the CAST study where flecainide increased rather than decreased mortality (Echt et al. 1991). The concept was that the use and development of anti-arrhythmic drugs should be based on insights in arrhythmia mechanisms, in ion channel properties and in the mechanisms of action of drugs that modulate ion currents. The group convened regularly and published several papers over the next 10 years. These concepts still stand today and are important in guiding therapy and drug development (Rosen & Janse, 2010). What has, however, remained elusive is effective and safe anti-arrhythmic drugs for ventricular arrhythmias. Patients at risk for ventricular tachycardia and fibrillation after myocardial infarction form the largest population in need of preventive drug therapy. The best available treatment today is implantation of an ICD, with or without ablation and supportive therapy, with very limited anti-arrhythmic drug options (Priori et al. 2015; Dan et al. 2018). This article has focused on Edward Carmeliet's scientific contributions. His colleagues were also fortunate to receive much timeless advice from him including the following. (1) It is important to revisit unresolved questions with new technologies, for example as shown by the refinement of his work on Na+ channels from two-microelectrode voltage clamp, to patch clamp, and finally transgenic technology. (2) Keep an open mind and be suspicious and critical of one's observations, using data to test rather than confirm hypotheses. It is noteworthy that in early papers in The Journal of Physiology, the abstract does not contain conclusions, only observations. (3) Embrace curiosity. Gathering data without preconceived ideas, often derided as a ‘fishing trip’, may prove its value at a time when we have enormous potential for unbiased big data. This allows curiosity-driven research and encourages us to pursue the unexpected. Finally, Edward Carmeliet showed us that science and curiosity are a powerful antidote for ageing. At a meeting in 2018 he explained how he had studied the biophysics and working of his hearing aid, which he now could tune optimally for lectures as well as for conversations in a crowd. He was a great scientist, mentor and friend.