![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
CELLULAR AND MOLECULAR
Department of Cell Physiology and Pharmacology, University of Leicester, Leicester, United Kingdom
Received for publication
August 31, 2005
Accepted
October 13, 2005.
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
) or hERG channel block by a variety of medications (Roden, 1998) can cause long QT syndrome. The type of drugs that block hERG are functionally and structurally diverse and include class III antiarrhythmics such as ibutilide, antibiotics such as erythromycin, antihistamines such as terfenadine, and antipsychotics such as sertindole. (Taglialatela et al., 1998; Kass and Cabo, 2000; Fermini and Fossa, 2003). This pharmacological "promiscuity" of hERG is because of, in part, the channel having an unusually large inner cavity capable of binding and trapping diverse compounds (Mitcheson et al., 2000). The inner helix (S6) of hERG lacks the Pro-X-Pro motif that in most other voltage-gated potassium channels is thought to introduce a sharp bend in the S6 subunit and consequently relatively small inner cavity (del Camino et al., 2000). In addition, hERG is unique in that it has two aromatic amino acid residues (Phe656 and Tyr652) per subunit that face into the cavity and allow hERG to form strong electrostatic as well as hydrophobic interactions with drugs. Mutation of Phe656 and Tyr652 to alanine has been shown to reduce sensitivity of hERG to a number of high- and low-affinity blockers (for examples, see Mitcheson et al., 2000; Perry et al., 2004; Witchel et al., 2004). Polar residues close to the K+ ion selectivity filter are also important for high-affinity block by some compounds (Perry et al., 2004).
The hERG channel is also expressed in neuronal and tumor tissue. Studies show hERG is expressed in cancerous tissues of various origins, including human endometrial (Cherubini et al., 2000) and human prolactin secreting adenocarcinomas (Bauer et al., 2003). hERG may influence tumor cell proliferation and invasiveness (Wang et al., 2002; Smith et al., 2003), but the mechanisms involving hERG channels remain largely unknown. Ether-a-go-go-related gene channels are expressed in a number of regions of the central nervous system, including the hippocampus, hypothalamus, and cerebellum (Papa et al., 2003) and may be an important component of M-like currents in the brain (Meves et al., 1999; Selyanko et al., 2000). Ether-a-go-go-related gene currents are modulated by many signal transduction pathways and regulation by G protein-coupled receptors may be important in controlling cellular excitability.
Elevations in cytosolic calcium play an important role in regulating numerous cellular processes, including excitation-contraction coupling in cardiac tissue (for review, see Berridge et al., 2000) and programmed cell death (for review, see Gupta and Pushkala, 1999). Caffeine is widely used either to elevate cytosolic calcium levels or in combination with calcium ATPase inhibitors to deplete intracellular stores; revealing relationships between the release of intracellular calcium and its physiological role. Caffeine is useful because it is highly membrane-permeable, induces rapid calcium release from intracellular stores, and is rapidly reversible. It increases cytosolic calcium by shifting the calcium dependence of the ryanodine receptor to low concentrations, thus leading to a release of calcium from intracellular stores under resting physiological intracellular calcium levels (Rousseau et al., 1988). Caffeine also has calcium-independent actions. It is a known phosphodiesterase inhibitor, increasing cytosolic cAMP levels and activating protein kinase A (PKA) (Butcher and Sutherland, 1962; Lindaman et al., 2002). A number of studies have suggested caffeine may act as a blocker of ion channels, for example, inward rectifier potassium channels in the GH3 rat pituitary cell line (Barros et al., 1996), the inward rectifier K+ current (IK1) and transient outward K+ current (Ito) in rat ventricular myocytes (Chorvatova and Hussein, 2003), and both outwardly and inwardly rectifying K+ channels in taste receptors (Zhao et al., 2002).
In this study, we report that caffeine inhibits the hERG potassium channel current and investigate its mechanism of action. Our data suggest that caffeine acts directly on hERG channels expressed in both a mammalian system and Xenopus oocytes. We have shown caffeine acts as a rapid open channel blocker and binds within the inner vestibule of the channel. These data may have implications for interpretation of data when using caffeine on cell and tissue preparations expressing the hERG channel.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
). Stage V and VI oocytes were injected with 5 to 15 ng of cRNA to wild type (WT), F656A and Y652A hERG, and recordings were made 1 to 4 days later. Oocytes were maintained at 18°C in Super Barth's solution containing 88 mM NaCl, 1 mM KCl, 0.41 mM CaCl2, 0.33 mM Ca(NO3)2, 1 mM MgSO4, 2.4 mM NaHCO3, 10 mM HEPES, 1 mM pyruvate, and 50 µg ml-1 gentamicin, pH 7.4 (adjusted with NaOH).
Electrophysiology. hERG expression in HEK293 cells was measured using the whole cell patch-clamp technique. Recordings were carried out with an Axopatch 200B amplifier (Molecular Devices, Sunnyvale, CA). Membrane currents were low pass filtered at 1 kHz and sampled at 2.5 kHz with a Digidata 1322 data acquisition system (Molecular Devices). Cells were perfused with extracellular Tyrode's solution, containing 140 mM NaCl, 1 mM MgCl2, 4 mM KCl, 10 mM glucose, 5 mM HEPES, and 2 mM CaCl2, pH 7.4 (adjusted using NaOH). Recordings were made at 35-37°C. A solenoid valvebased switching system was used to fully exchange solutions in the recordings chamber within 60 s. All drug solutions were made up daily to the required concentrations in extracellular Tyrode's solution. Borosilicate glass pipettes (Harvard Apparatus Inc., Holliston, MA) were pulled and fire polished to get final resistances of 2 to 5 M
. These were filled with an internal recording solution containing 130 mM KCl, 5 mM MgATP, and 10 mM HEPES, pH 7.2 (adjusted using KOH). In some experiments, 5 mM BAPTA-free acid was added, and the pH of the internal solution was readjusted. Whole cell capacitance measurements ranged between 8 and 29 pF, and series resistances were between 2 and 8 M and were compensated by 60 to 85%.
Membrane currents in Xenopus oocytes were recorded with the two-electrode voltage-clamp technique using a GeneClamp 500B amplifier (Molecular Devices) and Digidata 1322 (Molecular Devices). Data acquisition was performed with Clampex 8.1 (Molecular Devices). Cells were perfused with a low-chloride solution containing 96 mM Na-MES, 2 mM K-MES, 2 mM Ca(MES)2, 5 mM HEPES, and 1 mM MgCl2, pH 7.6 (adjusted with NaOH). Recordings were obtained at 20-21°C. Microelectrodes were filled with 3 M KCl, and the tips were broken to give resistances of 1 to 2 M.
Measurement of Cytosolic Calcium Levels. To monitor cytosolic calcium, hERG-HEK cells were loaded with 2 µM Fura-2-AM (Invitrogen) for 20 min at room temperature. In some experiments, cells were incubated with BAPTA-AM at the same time. The experiments were performed in nonvoltage-clamped cells to avoid dialysis of intracellular solutes via the patch pipette. Fluorescence was excited with light from a monochromator at 340 and 380 nm, and emitted light at >520 nm was imaged at 2-s intervals with a chargecoupled device camera (Photon Technology International, Lawrenceville, NJ). Data acquisition was performed using PTI Image-Master software to produce a 340/380 ratio of Fura-2 emission intensities. Experiments were conducted at 35 ± 1°C.
Voltage Protocols and Data Analysis. Electrophysiological recordings were analyzed using Clampfit (Molecular Devices) and Prism 3.0 (GraphPad Software Inc., San Diego, CA). To measure the effects of caffeine and other reagents on hERG currents hERG-HEK cells were held at a membrane potential of -80 mV, and 5-s depolarizations to 0 mV were applied. Tail currents were evoked by a 2- or 3-s repolarization to -50 mV. This protocol was repeated every 15 s to give sufficient time for current deactivation between test pulses. Cells were repetitively pulsed in control Tyrode's solution until current amplitudes stabilized before reagents were applied. Experiments to determine the caffeine binding site were done in Xenopus oocytes. For these experiments, cells were held at a membrane potential of -90 mV and depolarized to 0 mV for 5 s before being repolarized to -70 mV for 400 ms. This voltage protocol was repeated every 6 s to ensure that the channels were open (and inactivated) and that the inner cavity was accessible to caffeine for the majority of the time. For currents measured in both HEK cells and oocytes, peak tail current amplitudes were measured for each test pulse and normalized to currents before compound application. To analyze the voltage dependence of hERG channel activation in HEK cells, cells were held at a membrane potential of -80 mV, and depolarizations to potentials between -40 and +40 mV were applied in 10-mV increments for 5 s. Tail currents were evoked by a 2- or 3-s repolarization to -50 mV. In this protocol and all others in HEK cells, the start-to-start test pulse interval was 15 s. Peak tail current amplitudes were measured and normalized to the maximal amplitude, plotted against test pulse potential and fitted with a Boltzmann function, to obtain values for half-maximal activation (V0.5,act) and slope factors. To measure steady-state inactivation of hERG, a three-pulse voltage protocol was used. Membrane potential was depolarized to +40 mV for 500 ms from a holding potential of -80 mV. A 5-ms voltage pulse to potentials between -140 and +20 mV was applied, and current through noninactivated channels was measured at the beginning of a third pulse to +40 mV. Peak current at the beginning of the third pulse was normalized to maximum amplitude, plotted against the voltage of the preceding pulse, and fitted with a Boltzmann function to obtain values for V0.5,inact and slope factors. The "envelope of tails" voltage protocol was used to measure the time dependence of hERG block by caffeine. Cells were held at -80 mV, and depolarizations to 0 mV were initially applied for 20 ms. This interval was increased in 20-ms increments to a maximum of 400 ms. After each depolarization, current activation was monitored with repolarizing pulses to -50 mV for 60 ms to elicit tail currents, which were measured and normalized to the maximum tail current in control conditions, and plotted as a function of the test pulse duration. Statistical analysis was by unpaired Student's t tests unless indicated differently with significance accepted at p < 0.05.
|
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
The decrease of hERG current was seen at all potentials (Fig. 1C), and there was no significant shift in the voltage dependence of activation (Fig. 1D). V0.5,act and slope factor values in control conditions were -22.3 ± 1.9 mV and 7.8 ± 0.3 mV, respectively, compared with -23.1 ± 2.3 mV and 7.8 ± 0.4 mV (n = 6) in the presence of caffeine. The voltage dependence of inactivation was also not significantly affected by caffeine application. V0.5,inact values in control conditions were -73.7 ± 0.3 mV compared with -73.8 ± 1.0 mV in the presence of caffeine (n = 6; data not shown). Washout of caffeine was rapid. Currents began to recover on the first pulse after switching to control Tyrode's solution, with currents recovering to 99.1 ± 1.6% of control current within 30 s after the switch.
|
The Suppression of hERG Currents by Caffeine Is Not Due to Elevation of Cytosolic Calcium. To investigate whether the actions of caffeine were because of an increase in cytoplasmic calcium, the time course of calcium and hERG current changes were compared in parallel experiments. Cytoplasmic calcium was monitored with Fura-2. The Fura-2 ratio increased slowly to a mean value of 2.0 ± 0.2 after 4 min of caffeine application from a mean baseline value of 0.6 ± 0.2 (n 
12). Washout of the effects of caffeine also took more than 4 min. The elevation of calcium was abolished by extracellular application of 50 µM BAPTA-AM (Fig. 2A), a membrane-permeant calcium chelator. The slow changes of calcium are in contrast to calcium transients measured with the same apparatus in isolated rat cardiac myocytes, which peak in less than 5 s and decay in less than 20 s (Dr. G. Rodrigo, personal communication). This suggests the density of ryanodine receptors is low in this HEK293 cell line. Clearly, the time courses of calcium elevation and hERG current suppression were different; suggesting the hERG current response to caffeine is not mediated by cytosolic calcium. To further investigate whether the effects seen with caffeine are because of an increase in cytosolic calcium, voltage-clamped cells were dialyzed via the patch pipette with a solution containing 5 mM BAPTA. Cells were left for 10 min after attaining the whole cell configuration to allow complete dialysis of BAPTA into the cell. Application of caffeine in these conditions continued to cause a rapid attenuation of current to 53.9 ± 5.5% of control (n = 5; Fig. 2B) that was rapidly reversed upon caffeine washout, providing further evidence that caffeine does not exert its effects by an increase in calcium. Dialysis of cells with intracellular BAPTA depletes calcium stores by buffering released calcium and preventing replenishment. BAPTA (5 mM) should be far in excess of that needed to buffer calcium because 50 µM BAPTA-AM was sufficient to abolish the calcium transient. Nevertheless, to experimentally test whether the hERG response to caffeine could be because of incomplete calcium buffering by BAPTA, we measured the hERG current response to a second application. If the effect of caffeine was calcium-dependent, we would expect a greatly reduced response with a second application, because calcium stores would be significantly depleted. However, this was not the case. hERG current was reduced to 62.63 ± 2.05% of control with the second application (n = 5), which was not significantly different from the first application (Fig. 2C). These results provide strong evidence that caffeine does not act on hERG currents via a direct action of calcium on the channel.
Caffeine-Induced Effects on hERG Currents Are Not PKC- or cAMP-Dependent. Several studies suggest that hERG currents can be attenuated by protein kinase C (PKC), and this modulation can be blocked by a PKC-selective inhibitor bisindolylmalemide-1 (bis-1) (Thomas et al., 2003). To examine whether caffeine was exerting its effects via a PKC-dependent pathway, caffeine was applied to cells superfused for 3 min with 300 nM bis-1, a concentration that we have previously shown to be sufficient to inhibit PKC in these cells (S. L. Cockerill, A. B. Tobin, G. B. Willars, N. B. Standen, and J. S. Mitchenson, manuscript in preparation). Bis-1 (300 nM) had a minor influence on hERG current amplitude, with current amplitudes 95.9 ± 1.6% of control. Caffeine was then coapplied with bis-1. Bis-1 failed to reduce the caffeine response. Current amplitudes during caffeine application were 51.3 ± 5.5% of control (n = 6; Fig. 2D), which was not significantly different from the response in the absence of bis-1. Interestingly, attenuation of the hERG current upon caffeine application was slower in the presence of bis-1 compared with control, with maximal effects of caffeine being reached within 105 s, compared with 30 s in control conditions. The effects of caffeine in the presence of bis-1 were reversible upon washout, although recovery was also slower, reaching 90.9 ± 1.9% of control after 2 min. These results suggest that stimulation of PKC is not responsible for the attenuation of current by caffeine.
Caffeine is also a phosphodiesterase inhibitor, increasing the cytosolic cAMP levels, and therefore activating PKA (Butcher and Sutherland, 1962; Lindaman et al., 2002). Effects of both cAMP and PKA on the hERG channel have been documented previously (Cui et al., 2000, 2001), so we determined whether caffeine exerts its effects through either of these mechanisms. Forskolin, an adenylyl cyclase agonist, raises cAMP and causes slow, nonreversible attenuation of hERG current under our experimental conditions. The time course of action of forskolin was slower than caffeine (Fig. 2E), and forskolin application reduced current amplitudes to 75.9 ± 0.7% of control, which is less than caffeine. If cAMP levels are already high, it is unlikely that caffeine-induced phosphodiesterase inhibition would further modulate current. However, 5 mM caffeine continued to cause a rapid attenuation of current (to 65.14 ± 4.75% of control; n = 5; Fig. 2E) that also largely reversed upon washout. Thus, caffeine still attenuates hERG current in the presence of high cytosolic levels of cAMP. Together, our results suggest that caffeine does not affect the hERG channel current amplitude by calcium-, PKC-, PKA-, or cAMP-dependent pathways.
Open Channel Block and Trapping of Caffeine within the Inner Cavity of hERG. The rapid changes in current amplitude upon caffeine application and washout suggest caffeine may directly inhibit hERG currents. This possibility was explored with voltage protocols to investigate state-dependent block of hERG. Most blockers of hERG exhibit open state-dependent block and are thought to bind within the inner cavity of the channel (Mitcheson et al., 2000; Perry et al., 2004; Witchel et al., 2004; Sanguinetti and Mitcheson, 2005). The rapid action of caffeine on hERG currents could indicate either rapid open state block or closed state block. To distinguish between these possibilities, we used the envelope of tails voltage protocol, which allows the amount of block to be measured as pulse duration is increased (Suessbrich et al., 1997; Teschemacher et al., 1999). If caffeine is an open channel blocker, then inhibition will increase with progressively longer depolarizations. Voltage pulses to 0 mV from a holding potential of -100 mV were applied with durations of between 20 and 400 ms. Tail currents were elicited by a pulse to -50 mV for 60 ms. The protocol was carried out in control conditions, and repeated after 5 mM caffeine had been applied for 2 min. Mean results for tail currents normalized to control following the 400-ms pulse are shown in Fig. 3A. Caffeine decreased hERG current amplitudes in a pulse duration manner. The differences in current amplitudes before and during caffeine application were measured, and a mean percentage of inhibition was calculated and plotted against test pulse duration. Caffeine caused only a 2.7 ± 1.1% block with a 20-ms test pulse, which increased to 42.5 ± 0.8% after 400 ms, with a mean time constant for onset of current inhibition (
on) of 95.8 ± 21.0 ms (Fig. 3C). Similar experiments with an IC50 concentration (10 nM) of cisapride, a well known open channel blocker, demonstrated inhibition accumulated with channel activation, with a on of 383 ms (Walker et al., 1999). Thus, caffeine, like cisapride and many other long QT compounds, is an open channel blocker of hERG channels.
|
The above-mentioned results suggest the binding site for caffeine is likely to be within the inner cavity, behind the activation gate of the hERG channel. The drug binding site within the inner cavity is formed by amino acid residues from the inner helices (S6) pore helices. In particular, two aromatic residues on S6, Phe656 and Tyr652, are important for high-affinity block by nearly all hERG channel blockers that have been tested so far. Mutation of both residues to alanine dramatically reduces block (for review, see Recanatini et al., 2005; Sanguinetti and Mitcheson, 2005). To test whether caffeine also binds within the inner cavity, inhibition of Y652A and F656A hERG by caffeine was measured in Xenopus oocytes. This is an expression system that has been routinely used for investigating structural determinants of hERG channel block. WT currents were recorded by pulsing to 0 mV for 5 s from a holding potential of -90 mV. Tail currents were elicited by pulsing to -70 mV for 400 ms. This protocol was repeated every 6 s, so that channels were in the open state for most of the time. Attenuation of current in oocytes by 5 mM caffeine was similar to that seen in HEK cells (rapid and fully reversible), although percentage of block was reduced to 77.8 ± 2.4% of control (n = 7; Fig. 4A). Reduction of block in oocytes compared with mammalian cells has been frequently reported and is likely to exist because of accumulation of the drug in the lipophilic yolk of oocytes. Y652A currents showed a significantly reduced sensitivity to caffeine, with peak tail current during caffeine application only 92.6 ± 1.4% of control current (n = 9, Fig. 4C). F656A hERG currents were recorded with similar voltage protocols except that tail currents were measured at -140 mV to elicit larger amplitude currents (Mitcheson et al., 2000). F656A channels also showed a significantly lowered sensitivity to block by caffeine. Peak tail current during caffeine application was 93.6 ± 1.4% of control (n = 9; Fig. 4D). Thus, both F656A and Y652A decrease hERG current inhibition by caffeine, providing further evidence that caffeine exerts its effects on hERG currents by open state binding to residues in the inner cavity causing block rather than via a mechanism involving second messenger modulation of channel activity.
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Caffeine releases calcium into the cytosol from the ryanodine-sensitive calcium stores (Rousseau et al., 1988) and also increases cAMP levels (Butcher and Sutherland, 1962; Lindaman et al., 2002). Caffeine attenuated hERG channel current when calcium was both elevated by ionomycin (data not shown) and when buffered to low concentrations with BAPTA. Inhibition of PKC failed to alter attenuation of hERG channel currents by caffeine, and caffeine continued to reduce hERG current when cAMP was raised with forskolin. Thus, caffeine does not exert its effects by calcium- or cAMP-mediated pathways. Instead, our results suggest that caffeine directly blocks the hERG channel. Caffeine had rapid onset and recovery kinetics, and inhibition was highly state-dependent. Inhibition increased progressively as pulse duration was lengthened and channels became activated. Recovery from block was also more complete with longer duration test pulses. These results indicate that block in the open and/or inactivated state is the mechanism by which caffeine suppresses hERG currents.
Inhibition of WT channels expressed in HEK293 cells and Xenopus oocytes was very similar, despite the gross differences in size and subcellular organization of the two cell types. This provides further evidence that caffeine has a direct rather than modulatory effect on hERG channels. To locate the binding site of caffeine, we used two mutants of the hERG channel, F656A and Y652A, that are known to reduce the sensitivity of hERG to a large number of long QT-inducing drugs that are open channel blockers. Both are aromatic residues that face into the inner cavity of hERG and can mediate hydrophobic and electrostatic interactions with structurally diverse compounds. The 
electron faces of aromatic residues can mediate -stacking interactions with aromatic moieties on compounds, and cation- interactions with charged amines. We found that replacing either the Phe656 or Tyr652 residues with the small amino acid alanine reduced the sensitivity of the channel to caffeine. This suggests caffeine interacts with both residues in the WT channel and therefore binds within the inner cavity of the channel.
While investigating whether the caffeine response was mediated by PKC, we found inhibition of hERG by caffeine to be slower in the presence of the PKC inhibitor bis-1. We and others (Thomas et al., 2004) have shown bis-1 to directly inhibit hERG current at concentrations higher than 300 nM. Therefore, 300 nM bis-1, although not blocking the channel, may be competing with caffeine and partially obstructing caffeine entry into the hERG channel, leading to a slower block of the channel. Recovery of caffeine inhibition with washout in the presence of bis-1 was also slower and may also be because of bis-1 hindering caffeine exit from the channel.
Caffeine is an alkaloid. The core of the molecule is a heterocyclic purine that is able to interact with aromatic residues on proteins. Caffeine can bind to the purine inhibitor site on glycogen phosphorylase A (Tsitsanou et al., 2000). This binding occurs by -stacking interactions in which caffeine intercalates between two aromatic rings, Phe285 and Tyr613. Kapuscinski et al. (2002) found that caffeine binds to aromatic mutagens, such as quinacrine mustard, and reduces their pharmacological activity. Thus, caffeine may act as a protective agent in some cancer types. The interaction between caffeine and the aromatic agents was again because of -stacking interactions. Caffeine can decrease the effective concentrations of polycyclic aromatic agents but not aliphatic agents (Piosik et al., 2003). These three studies, although not carried out on ion channels, demonstrate the ability of caffeine to interact with aromatic residues. Our results are consistent with -stacking interactions between caffeine and the Phe656 and Tyr652 residues on hERG, although other interactions with inner cavity residues may also take place.
Several studies have shown that caffeine can inhibit ion channels. The rapid, reversible effects of caffeine shown in our study concur with results of Barros et al. (1996), who showed caffeine to cause a reversible increase of action potential firing frequency along with a depolarization of membrane potential in GH3 cells, a rat anterior pituitary cell line. They were unable to mimic the effects of caffeine with forskolin and 3-isobutyl-1-methylxanthine, which both raise cAMP levels. The effects of caffeine could not be inhibited when cells were incubated with ryanodine or thapsigargin, to deplete intracellular calcium stores, or by buffering cytosolic calcium with EGTA dialyzed into cells via the patch pipette. However, caffeine significantly inhibited an inwardly rectifying potassium current endogenously expressed in these cells. It has since been shown that GH3 cells express channels with sequence homology to hERG that are blocked by astemizole (Barros et al., 1997). Thus, it is likely that the caffeine-sensitive current is an endogenously expressed rat ERG channel.
Other membrane currents reported to be inhibited by caffeine include IK1 and Ito (Chorvatova and Hussein, 2003). In rat ventricular myocytes, caffeine partially inhibited the inward rectifier current IK1 in a concentration-dependent manner. Caffeine also inhibited Ito without altering the voltage and time dependence of activation or inactivation and in a fully reversible manner. The same study also showed 10 mM caffeine to partially inhibit the delayed rectifier IK, although selectivity for the rapid or slow components was not established. GABAA current in hippocampal slices is also significantly inhibited by caffeine (Taketo et al., 2004), in a manner that could not be inhibited by calcium store depletion, or inclusion of EGTA or BAPTA in the pipette solution, and could not be mimicked by ryanodine or forskolin, indicating the inhibition of current may be because of direct block of channels.
Therefore, several studies now suggest that ion channel block needs to be taken into consideration when using caffeine as a calcium-release tool and that it is not specific for ryanodine receptors. At present, caffeine is widely used to investigate calcium-dependent transduction pathways in cellular preparations. For example, it is frequently used to study excitation-contraction coupling in cardiac myocyte preparations (Seidler et al., 2003; Gomez et al., 2004; Huang et al., 2005). Many studies use caffeine at concentrations higher than 5 mM, which we found to profoundly inhibit hERG currents. Thus, caffeine may not only elicit calcium release but also inhibit IKr, Ito, and IK1 to significantly alter action potential repolarization in nonvoltage-clamped cells.
hERG is expressed in a variety of tissues in addition to cardiac myocytes, including some cancer cells (Cherubini et al., 2000) and neuronal tissues (Papa et al., 2003). Our results indicate that care should be taken with interpretation of results when using caffeine in hERG-expressing excitable cells.
Is there a physiological relevance to caffeine inhibition of hERG channels? Caffeine is a common dietary constituent in many of today's cultures. However, it is unlikely to be a long QT syndrome-inducing drug because it only produces a significant inhibition of hERG channel current when used in millimolar concentrations. A study carried out in 1982 showed the average plasma concentration of caffeine in non-smoking males was 10.9 µM l-1 (Smith et al., 1982), a concentration that would not significantly block hERG. It seems extremely unlikely that caffeine would cause any significant change to the ECG as a result of blocking the hERG channel.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
ABBREVIATIONS: hERG, human ether-a-go-go-related gene; PKA, protein kinase A; HEK, human embryonic kidney; WT, wild type; MES, 2-(N-morpholino)ethanesulfonic acid; AM, acetoxymethyl ester; PKC, protein kinase C; bis-1, bisindolylmalemide-1; BAPTA, 1,2-bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid.
Address correspondence to: Dr. John Mitcheson, Department of Cell Physiology and Pharmacology, Maurice Shock Medical Sciences Building, University Rd., University of Leicester, Leicester LE1 9HN, United Kingdom. E-mail: jm109{at}leicester.ac.uk
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Barros F, Del Camino D, Pardo L, and De la Pena P (1996) Caffeine enhancement of electrical activity through direct blockade of inward rectifying K+ currents in GH3 rat anterior pituitary cells. Pflueg Arch Eur J Physiol 431: 443-451.[CrossRef][Medline]
Barros F, Del Camino D, Pardo L, Palomero T, Giraldez T, and De la Pena P (1997) Demonstration of an inwardly rectifying K+ current component modulated by thyrotropin-releasing hormone in GH3 rat anterior pituitary cells. Pflueg Arch Eur J Physiol 435: 119-129.[CrossRef][Medline]
Bauer C, Wulfsen I, Schafer R, Glassmeier G, Wimmers S, Flitsch J, Ludecke D, and Schwartz J (2003) HERG K+ currents in human prolactin-secreting adenoma cell. Pflueg Arch Eur J Physiol 445: 589-600.[Medline]
Berridge M, Lipp P, and Bootman M (2000) The versatility and universality of calcium signaling. Nat Rev Mol Cell Biol 1: 11-21.[CrossRef][Medline]
Butcher R and Sutherland E (1962) Adenosine 3',5'-phosphate in biological materials. I. Purification and properties of cyclic 3',5'-nucleotide phosphodiesterase and use of this enzyme to characterize adenosine 3',5'-phosphate in human urine. J Biol Chem 237: 1244-1250.
Cherubini A, Taddel G, Crociani O, Paglierani M, Buccoliero A, Fontana L, Noci I, Borri P, Borrani E, Giachi M, et al. (2000) HERG potassium channels are more frequently expressed in human endometrial cancer as compared to non-cancerous endometrium. Br J Cancer 83: 1722-1729.[CrossRef][Medline]
Chorvatova A and Hussein M (2003) Effects of caffeine in potassium currents in isolated rat ventricular myocytes. Pflueg Arch Eur J Physiol 446: 422-428.[CrossRef][Medline]
Cui J, Kagan A, Qin D, Mathew J, Melman Y, and McDonald T (2001) Analysis of the cyclic nucleotide binding domain of the HERG potassium channel and interactions with KCNE2. J Biol Chem 276: 17244-17251.
Cui J, Menman Y, Palma E, Fishman G, and McDonald T (2000) Cyclic AMP regulates the HERG K channel by dual pathways. Curr Biol 10: 671-674.[CrossRef][Medline]
del Camino D, Holmgren M, Liu Y, and Yellen G (2000) Blocker protection in the pore of a voltage-gated K+ channel and its structural implications. Nature (Lond) 403: 321-325.[CrossRef][Medline]
Fermini B and Fossa A (2003) The impact of drug-induced QT interval prolongation on drug discovery and development. Nat Rev Drug Discov 2: 439-447.[CrossRef][Medline]
Gomez AM, Schuster I, Fauconnier J, Prestle J, Hassenfuss G, and Richard S (2004) FKBP12.6 overexpression decreases Ca2+ spark amplitude but enhances [Ca2+]i transient in rat cardiac myocytes. Am J Physiol 287: H1987-H1993.
Gupta PD and Pushkala K (1999) Importance of the role of calcium in programmed cell death: a review. Cytobios 99: 89-93.
Huang J, Hove-Madsen L, and Tibbits GF (2005) Na+/Ca2+ exchange activity in neonatal rabbit ventricular myocytes. Am J Physiol 288: C195-C203.
Kapuscinski J, Ardelt B, Piosik J, Zdunek M, and Darzynkiewicz Z (2002) The modulation of the DNA-damaging effect of polycyclic aromatic agents by xanthines. Part I. Reduction of cytostatic effects of quinacrine mustard by caffeine. Biochem Pharmacol 63: 625-634.[CrossRef][Medline]
Kass R and Cabo C (2000) Channel structure and drug-induced cardiac arrhythmia Proc Natl Acad Sci USA 97: 11683-11684.
Lindaman B, Hinkhouse M, Conklin J, and Cullen J (2002) The effect of phosphodiesterase inhibition on gallbladder motility in vitro. J Surg Res 105: 102-108.[CrossRef][Medline]
Meves H, Schwarz JR, and Wulfsen I (1999) Separation of M-like current and ERG current in NG108-15 cells. Br J Pharmacol 127: 1213-1223.[CrossRef][Medline]
Mitcheson J, Chen J, and Sanguinetti M (2000) Trapping of a methanesulfonamide by closure of the HERG potassium channel activation. J Gen Physiol 115: 229-240.
Mitcheson JS, Chen J, Lin M, Culberson C, and Sanguinetti MC (2000) A structural basis for drug-induced long QT syndrome. Proc Natl Acad Sci USA 97: 12329-12333.
Papa M, Boscia F, Canitano A, Castaldo P, Sellitti S, Annunziato L, and Taglialatela M (2003) Expression pattern of the ether-a-go-go-related (ERG) K+ channel-encoding genes ERG1, ERG2 and ERG3 in the adult rat central nervous system. J Comp Neurol 466: 119-135.[CrossRef][Medline]
Perry M, de Groot M, Helliwell R, Leishman D, Tristani-Firouzi M, Sanguinetti M, and Mitcheson J (2004) Structural determinants of HERG channel block of clofilium and ibutilide. Mol Pharmacol 66: 240-248.
Piosik J, Ulanowska K, Gwizdek-Wisniewska A, Czy A, Kapuscinski J, and Wegrzyn G (2003) Alleviation of mutagenic effects of polycyclic aromatic agents (quinacrine mustard, ICR-191 and ICR-170) by caffeine and pentoxifylline. Mutat Res 530: 47-57.[Medline]
Recanatini M, Poluzzi E, Masetti M, Cavalli A, and De Ponti F (2005) QT prolongation through HERG K(+) channel blockade: current knowledge and strategies for the early prediction during drug development. Med Res Rev 25: 133-166.[CrossRef][Medline]
Roden DM (1998). Mechanisms and management of proarrhythmia. Am J Cardiol 82: 49I-57I.[CrossRef][Medline]
Rousseau E, Ladine J, Liu Q, and Meissner G (1988) Activation of the Ca2+ release channel of skeletal muscle sarcoplasmic reticulum by caffeine and related compounds. Arch Biochem Biophys 267: 75-86.[CrossRef][Medline]
Sanguinetti M, Curran M, Spector P, and Keating M (1996) Spectrum of HERG K-channel dysfunction in an inherited cardiac arrhythmia. Proc Natl Acad Sci USA 93: 2208-2212.
Sanguinetti MS and Mitcheson JS (2005) Predicting drug-HERG channel interactions that cause acquired long QT syndrome. Trends Pharmacol Sci 26: 119-124.[CrossRef][Medline]
Seidler T, Miller SL, Loughrey CM, Kania A, Burow A, Kettlewell S, Teucher N, Wagner S, Kogler H, Meyers MB, et al. (2003) Effects of adenovirus-mediated sorcin overexpression on excitation-contraction coupling in isolated rabbit cardiomyocytes. Circ Res 93: 132-139.
Selyanko AA, Hadley JK, Wood IC, Abogadie FC, Jentsch TJ, and Brown DA (2000) Inhibition of KCNQ1-4 potassium channels expressed in mammalian cells via M1 muscarinic acetylcholine receptors. J Physiol (Lond) 522: 349-355.
Smith G, Tsui H, Newell E, Jiang X, Zhu X-P, Tsui F, and Schlichter L (2003) Functional up-regulation of HERG K+ channels in neoplastic hematopoietic cells. J Biol Chem 277: 18528-18534.
Smith J, Pearson S, and Marks V (1982) Plasma caffeine concentration in outpatients. Lancet 2: 985-986.[Medline]
Suessbrich H, Schonherr R, Heinemann SH, Attali B, Lang F, and Busch AE (1997) The inhibitory effect of the antipsychotic drug haloperidol on HERG potassium channels expressed in Xenopus oocytes. Br J Pharmacol 120: 968-974.[CrossRef][Medline]
Taglialatela M, Castaldo P, Pannaccione A, Giorgio G, and Annunziato L (1998) Human ether-a-go go related gene (HERG) K channels as pharmacological targets: present and future implications. Biochem Pharmacol 55: 1741-1746.[CrossRef][Medline]
Taketo M, Matsuda H, and Yoshioka T (2004) Calcium-independent inhibition of GABA(A) current by caffeine in hippocampal slices. Brain Res 1016: 229-239.[CrossRef][Medline]
Teschemacher AG, Seward EP, Hancox JC, and Witchel HJ (1999) Inhibition of the current of heterologously expressed HERG potassium channels by imipramine and amitriptyline. Br J Pharmacol 128: 479-485.[CrossRef][Medline]
Thomas D, Hammerling BC, Wimmer AB, Wu K, Ficker E, Kuryshev YA, Scherer D, Kiehn J, Katus HA, Schoels W, et al. (2004) Direct block of HERG potassium channels by the protein kinase C inhibitor bisindolylmaleimide I (GF109203X). Cardiovasc Res 64: 467-476.
Thomas D, Zhang W, Wu K, Wimmer AB, Gut B, Wendt-Nordahl G, Kathofer S, Kreye V, Katus H, Schoels W, et al. (2003) Regulation of HERG potassium channel activation by protein kinase C independent of direct phosphorylation of the channel protein. Cardiovasc Res 59: 14-26.
Tsitsanou K, Skarnaki V, and Oikonomakos N (2000) Structural basis of the synergistic inhibition of glycogen phosphorylase a by caffeine and a potential antidiabetic drug. Arch Biochem Biophys 384: 245-254.[CrossRef][Medline]
Walker BD, Singleton CB, Bursill JA, Wyse KR, Valensuela SM, Qui MR, Breit SN, and Campbell TJ (1999) Inhibition of the human ether-a-go-go related gene (HERG) potassium channel by cisapride: affinity for open and inactivated states. Br J Pharmacol 128: 444-450.[CrossRef][Medline]
Wang H, Zhang Y, Cao L, Han H, Wang J, Yang B, Nattel S, and Wang Z (2002) HERG K+ channel, a regulator of tumour cell apoptosis and proliferation. Cancer Res 62: 4843-4848.
Witchel HJ, Dempsey CE, Sessions RB, Perry M, Milnes JT, Hancox JC, and Mitcheson JS (2004) The low-potency, voltage-dependent HERG blocker propafenone-molecular determinants and drug trapping. Mol Pharmacol 66: 1201-1212.
Zhao F-L, Lu S-G, and Herness S (2002) Dual actions of caffeine on voltage-dependent currents and intracellular calcium in taste receptor cells. Am J Physiol 283: R115-R129.
This article has been cited by other articles:
|
|
 |
|
  S. L. Cockerill, A. B. Tobin, I. Torrecilla, G. B. Willars, N. B. Standen, and J. S. Mitcheson Modulation of hERG potassium currents in HEK-293 cells by protein kinase C. Evidence for direct phosphorylation of pore forming subunits J. Physiol., June 1, 2007; 581(2): 479 - 493. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Mao, Y. Chai, and Y.-F. Lin Dual regulation of the ATP-sensitive potassium channel by caffeine Am J Physiol Cell Physiol, June 1, 2007; 292(6): C2239 - C2258. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
 |
 |
 |
