Many drugs inhibit the human ether-a-go-go-related gene (HERG) cardiac Kϩ channel. This leads to action potential prolongation on the cellular level, a prolongation of the QT interval on the electrocardiogram, and sometimes cardiac arrhythmia. To date, no activators of this channel have been reported. Here, we describe the in vitro electrophysiological effects of (3R,4R)-4-[3-(6-methoxyquinolin-4-yl)-3-oxo-propyl]-1-[3-(2,3,5-trifluoro-phenyl)-prop-2-ynyl]-piperidine-3-carboxylic acid (RPR260243), a novel activator of HERG. Using patch-clamp electrophysiology, we found that RPR260243 dramatically slowed current deactivation when applied to cells stably expressing HERG. The effects of RPR260243 on HERG channel deactivation were temperature-and voltagedependent and occurred over the concentration range of 1 to 30 M. RPR260243-modified HERG currents were inhibited by dofetilide (IC 50 ϭ 58 nM). RPR260243 had little effect on HERG current amplitude and no significant effects on steady-state activation parameters or on channel inactivation processes. RPR260243 displayed no activator-like effects on other voltagedependent ion channels, including the closely related erg3 K ϩ channel. RPR260243 enhanced the delayed rectifier current in guinea pig myocytes but, when administered alone, had little effect on action potential parameters in these cells. However, RPR260243 completely reversed the action potential-prolonging effects of dofetilide in this preparation. Using the Langendorff heart method, we found that 5 M RPR260243 increased T-wave amplitude, prolonged the PR interval, and shortened the QT interval. We believe RPR260243 represents the first known HERG channel activator and that the drug works primarily by inhibiting channel closure, leading to a persistent HERG channel current upon repolarization. Compounds like RPR260243 will be useful for studying the physiological role of HERG and may one day find use in treating cardiac disease.

The slower kinetics of insulin release from pancreatic islet beta cells, as compared with other regulated secretory processes such as chromaffin granule secretion, can in part be explained by the small number of the insulin granules that are docked to the plasma membrane and readily releasable. In type-2 diabetes, the kinetics of insulin secretion become grossly distorted, and, to therapeutically correct this, it is imperative to elucidate the mechanisms that regulate priming and secretion of insulin secretory granules. Munc13-1, a synaptic protein that regulates SNARE complex assembly, is the major protein determining the priming of synaptic vesicles. Here, we demonstrate the presence of Munc13-1 in human, rat, and mouse pancreatic islet beta cells. Expression of Munc13-1, along with its cognate partners, syntaxin 1a and Munc18a, is reduced in the pancreatic islets of type-2 diabetes non-obese Goto-Kakizaki and obese Zucker fa/fa rats. In insulinoma cells, overexpressed Munc13-1-enhanced green fluorescent protein is translocated to the plasma membrane in a temperature-dependent manner. This, in turn, greatly amplifies insulin exocytosis as determined by patch clamp capacitance measurements and radioimmunoassay of the insulin released. The potentiation of exocytosis by Munc13-1 is dependent on endogenously produced diacylglycerol acting on the overexpressed Munc13-1 because it is blocked by a phospholipase C inhibitor (U73122) and abrogated when the diacylglycerol binding-deficient Munc13-1 H567K mutant is expressed instead of the wild type protein. Our data demonstrate that Munc13-mediated vesicle priming is not restricted to neurotransmitter release but is also functional in insulin secretion, where it is subject to regulation by the diacylglycerol second messenger pathway. In view of our findings, Munc13-1 is a potential drug target for therapeutic optimization of insulin secretion in diabetes.

Aims/hypothesis. Syntaxin-1A (Syn-1A) is known to play a negative regulatory role in insulin secretion but the precise mechanisms for its action are not clear. Syn-2, ±3 and ±4 are also present in islet beta cells but their functions are not known. Here, we investigated the role of these syntaxins in the insulin secretory process. Methods. We examined the following effects of Syn-1, ±2, ±3 and ±4 expression in insulinoma beta-cell lines. Endogenous insulin secretion was measured by batch radioimmunoassay (RIA) and single cell patch clamp capacitance measurements. The l-type Ca 2+ channel activity was studied by patch clamp electrophysiology. Insulin gene transcription was examined by Northern blotting and measurement of insulin gene promoter activity by the co-expression of cyan fluorescent protein-labelled rat insulin promoter. Results. Syn-1A or ±3, but not Syn-2 or ±4 overexpression, inhibited K + -induced insulin release as determined by RIA (49.7 5.5 % and 49.1 6.2 %, respectively) and electrophysiologic membrane capacitance measurements (68.0 21.0 % and 58.0 13.2 %, respectively). Overexpressed Syn-1A and ±3, but not Syn-2, inhibited Ca 2+ channel current amplitude by 39.5 11.6 % and 52.7 6.0 %, respectively. Of note, overexpression of Syn-1A and ±3 also reduced single cell (by confocal microscopy) and total cellular endogenous insulin content (by RIA) by 24.8 4.2 % and 31.8 3.9 %, respectively. This correlated to a reduction in endogenous insulin mRNA by 24.5 4.2 % and 25.7 4.2 %, respectively. This inhibition of insulin biosynthesis is mainly at the level of insulin gene transcription as demonstrated by an inhibition of insulin gene promoter activity (53.3 9.15 % and 39.0 6.8 %, respectively). Conclusions/interpretation. These results demonstrate that Syn-1A and ±3 possess strong inhibitory actions on both insulin exocytosis and insulin biosynthesis whereas Syn-2 and ±4 do not inhibit the insulin secretory process. [Diabetologia (2002)

Cognate soluble N-ethylmaleimide-sensitive factor attachment protein receptor (SNARE) proteins are now known to associate the secretory vesicle with both the target plasma membrane and Ca 2؉ channels in order to mediate the sequence of events leading to exocytosis in neurons and neuroendocrine cells. Neuroendocrine cells, particularly insulin-secreting islet ␤-cells, t-SNARE proteins, 25-kDa synaptosomal-associated protein (SNAP-25), and syntaxin 1A, independently inhibit the L-type Ca 2؉ channel (L Ca ). However, when both are present, they actually exhibit stimulatory actions on the L Ca . This suggests that the positive regulation of the L Ca is conferred by a multi-SNARE protein complex. We hypothesized an alternate explanation, which is that each of these SNARE proteins possess distinct inhibitory and stimulatory domains that act on the L Ca . These

Tolterodine is a muscarinic antagonist widely used in the treatment of urinary incontinence. Although tolterodine has not been reported to alter cardiac repolarization, it is chemically related to other muscarinic antagonists known to prolong cardiac repolarization. For this reason, we studied the effects of tolterodine on cardiac ion channels and action potential recordings. Using patch-clamp electrophysiology, we found that tolterodine was a potent antagonist of the human ether-a-go-go-related gene (HERG) K ϩ channel, displaying an IC 50 value of 17 nM. This potency was similar to that observed for the antiarrhythmic drug dofetilide (IC 50 of 11 nM). Tolterodine block of HERG displayed a positive voltage dependence, suggesting an interaction with an activated state. Tolterodine had little effect on the human cardiac Na ϩ channel at concentrations of up to 1 M. Inhibition of L-type Ca 2ϩ currents by tolterodine was frequency-dependent with IC 50 values measuring 143 and 1084 nM at 1 and 0.1 Hz, respectively. Both tolterodine and dofetilide prolonged action potential duration in single guinea pig myocytes over the concentration range of 3 to 100 nM. However, prolongation was significantly larger for dofetilide compared with tolterodine. Tolterodine seems to be an unusual drug in that it blocks HERG with high affinity, but produces little QT prolongation clinically. Low plasma levels after therapeutic doses combined with mixed ion channel effects, most notably Ca 2ϩ channel blockade, may serve to attenuate the QT prolonging effects of this potent HERG channel antagonist.