Exp Physiol 89.3 pp 237–242 237 Experimental Physiology Differential expression of the mechanosensitive potassium channel TREK-1 in epicardial and endocardial myocytes in rat ventricle Joy H. C. Tan, Weihong Liu and David A. Saint School of Molecular and Biomedical Sciences, University of Adelaide, Adelaide, SA 5005, Australia Mechanoelectric feedback (MEF) is the process by which mechanical forces on the myocardium induce electrical responses. It is thought that MEF is important in controlling the beat to beat force of contraction in the ventricle, in response to fluctuations in load, and it may also play a role in controlling the dispersion of repolarization. The transduction mechanism for MEF is via stretch sensitive ion channels in the surface membrane of myocytes. Two types of stretch sensitive channels have been described; a non-selective cation channel, and a potassium selective channel. TREK-1 is a member of the recently cloned tandem pore potassium channels that has been shown to be mechanosensitive and to be expressed in rat heart. Here we report that the gene expression level of TREK-1, quantified using real-time RT-PCR against glyceraldehyde phosphate dehydrogenase (GAPDH) as a comparator gene, was found to be 0.34 ± 0.14 in endocardial cells compared to 0.02 ± 0.02 in epicardial cells (P < 0.05). To confirm that this is reflected in a different current density, whole cell TREK-1 currents, activated by chloroform, were recorded with patch clamp techniques in epicardial and endocardial cells. TREK-1 current density in epicardial and endocardial cells was 0.21 ± 0.06 pA/pF and 0.8 ± 0.27 pA/pF, respectively (P ≤ 0.05). We discuss the implications of this differential expression of TREK-1 for controlling action potential repolarization when the myocardium is stretched. We hypothesize that the gene expression of TREK-1 is controlled by the different amounts of stretch experienced by muscle cells across the ventricular wall. (Resubmitted 18 December 2003; accepted 9 January 2004; first published online 17 February 2004) Corresponding author D. A. Saint: School of Molecular and Biomedical Sciences, University of Adelaide, Adelaide, SA 5005, Australia. Email: david.saint@adelaide.edu.au It is thought that a common arrhythmogenic stimulus is temporal and spatial dispersion of repolarization of the ventricular action potential leading to local re-entrant circuits, and hence triggering after- depolarizations (e.g. Volders et al. 2000). Dispersion of repolarization is normally minimized by the different durations of epicardial and endocardial action potentials, the briefer action potentials in endocardial cells tending to synchronize repolarization in endocardial cells with that in epicardial cells, despite their later activation (Cowan et al. 1988). This difference in action potential duration is largely a result of epicardial and endocardial cells having different gene expression levels of potassium channels. For example, J. H. C. Tan and W. Liu contributed equally to this work. voltage-activated potassium channels such as Kv 4.2 and Kv LQT1 have been found to be expressed differentially across the ventricular wall (Dixon et al. 1996; Pereon et al. 2000), with the expression of Kv 4.2 being more than eight times higher in epicardial muscle cells compared to papillary muscle cells (Dixon & McKinnon, 1994). The observation that there are stretch sensitive, or ‘mechanosensitive’, ion channels in myocytes, at least one type being a potassium channel (Hu & Sachs, 1997), raises the possibility that mechanical forces on the myocardium may also influence action potential repolarization. Indeed, there is direct evidence for this in animals and humans (e.g. Greve et al. 2001; Ravelli et al. 1994). Hence, a logical extension of the idea that action potential repolarization is differentially controlled C  The Physiological Society 2004 DOI: 10.1113/expphysiol.2003.027052