• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of brjpharmLink to Publisher's site
Br J Pharmacol. Mar 2005; 144(6): 821–829.
Published online Jan 31, 2005. doi:  10.1038/sj.bjp.0706068
PMCID: PMC1576064

Inhibition of the human two-pore domain potassium channel, TREK-1, by fluoxetine and its metabolite norfluoxetine


  1. Block of the human two-pore domain potassium (2-PK) channel TREK-1 by fluoxetine (ProzacR) and its active metabolite, norfluoxetine, was investigated using whole-cell patch-clamp recording of currents through recombinant channels in tsA 201 cells.
  2. Fluoxetine produced a concentration-dependent inhibition of TREK-1 current that was reversible on wash. The IC50 for block was 19 μM. Block by fluoxetine was voltage-independent. Fluoxetine (100 μM) produced an 84% inhibition of TREK-1 currents, but only a 31% block of currents through a related 2-PK channel, TASK-3.
  3. Norfluoxetine was a more potent inhibitor of TREK-1 currents with an IC50 of 9 μM. Block by norfluoxetine was also voltage-independent.
  4. Truncation of the C-terminus of TREK-1 (Δ89) resulted in a loss of channel function, which could be restored by intracellular acidification or the mutation E306A. The mutation E306A alone increased basal TREK-1 current and resulted in a loss of the slow phase of TREK-1 activation.
  5. Progressive deletion of the C-terminus of TREK-1 had no effect on the inhibition of the channel by fluoxetine. The E306A mutation, on the other hand, reduced the magnitude of fluoxetine inhibition, with 100 μM producing only a 40% inhibition.
  6. It is concluded that fluoxetine and norfluoxetine are potent inhibitors of TREK-1. Block of TREK-1 by fluoxetine may have important consequences when the drug is used clinically in the treatment of depression.
Keywords: Two-pore domain potassium channel, TASK-3, TREK-1, fluoxetine, norfluoxetine, C-terminus, potassium current


Background or ‘leak' potassium (K) currents are present in almost all neurons and control the excitability and the resting membrane potential of these cells (Hille, 2001). It is becoming clear that the channel proteins underlying many of these leak K currents are members of the two-pore domain potassium (2-PK) channel family (Goldstein et al., 2001; O'Connell et al., 2002; Lesage 2003). Currently, there are known to be at least 16 mammalian channels in the 2-PK channel family. These can be divided, loosely, into six subfamilies on the basis of structural and functional similarities (Alexander et al., 2004). One of these subfamilies of 2-PK channels, the TREK subfamily, consists, at present, of three identified channel proteins, TREK-1, TREK-2 and TRAAK. In this study, we have considered the properties of one of these channels, TREK-1 (Fink et al., 1996), and its regulation by both the widely prescribed antidepressant agent, fluoxetine (ProzacR), and its major active metabolite, norfluoxetine.

TREK-1 channels are expressed throughout the CNS (Fink et al., 1996; Hervieu et al., 2001; Talley et al., 2001; Gu et al., 2002) with a complex, species-dependent distribution (see Meadows et al., 2000), but with notable levels of expression in the hippocampus, cerebellum, cerebral cortex (Hervieu et al., 2001) and hypothalamus (Maingret et al., 2000a). TREK-1 channels are regulated by a wide range of physiological and pharmacological mediators. Physiologically, this regulatory activity will mean that the proportion of active TREK-1 channels is constantly varying and, as a consequence, the excitability of the neuron they are expressed in, will also vary. For example, TREK-1 channels are activated by membrane stretch (Patel et al., 1998), arachidonic acid (Patel et al., 1998), intracellular acidification (Maingret et al., 1999), heat (Maingret et al., 2000a), lysophospholipids (Maingret et al., 2000b), activation of cGMP-dependent protein kinase (Koh et al., 2003), copper ions (Gruss et al., 2004b) and a number of anaesthetic agents (Patel et al., 1999; Gruss et al., 2004a). The cytosolic C-terminal domain of the channel is critical for many of these activations (Patel et al., 1998; Maingret et al., 1999; 2000a; Gruss et al., 2004a). In addition, TREK-1 channels are inhibited by the antipsychotic agent, chlorpromazine (Patel et al., 1998), and by a number of G-protein-coupled neurotransmitter receptors (e.g. Chemin et al., 2003).

A recent study using TREK-1 knockout mice (Heurteaux et al., 2004) has demonstrated the importance of TREK-1 channels for a number of primary CNS functions. TREK-1 channels have been shown to have an important role in neuroprotection against epilepsy and ischaemia. For example, the neuroprotective action of polyunsaturated fatty acids (Lauritzen et al., 2000; Blondeau et al., 2002) is lost in TREK-1 knockout animals (Heurteaux et al., 2004). Furthermore, these mice are resistant to anaesthesia by volatile anaesthetics. It follows that the use of any pharmacological agents that interfere with TREK-1 channel activity, such as we demonstrate here for fluoxetine, will have important consequences for CNS function.

Preliminary reports of some of these results have been published (Chumbley et al., 2003; Kennard et al., 2004).


Tissue culture

Modified HEK-293 cells (tsA 201) were maintained in 5% CO2 in a humidified incubator at 37°C in growth media (89% Dulbecco's modified Eagle's medium; 10% heat-inactivated foetal bovine serum; 1% penicillin (10,000 U ml−1) and streptomycin (10 mg ml–1). When the tsA cells were 80% confluent, they were split and plated for transfection onto glass coverslips coated with poly-D-lysine (1 mg ml−1) to ensure good cell adhesion. The tsA 201 cells were transiently transfected using the calcium phosphate method. cDNA (0.33 μg) encoding a TREK-1 or TASK-3 subunits was added to each 15 mm well, and 0.33 μg of a plasmid encoding the cDNA of green fluorescent protein was included to identify cells expressing two-pore domain channels. Following a 24 h incubation period at 3% CO2, the cells were rinsed with saline and fresh growth medium was added to the wells. The cells were incubated at 37°C with 5% CO2 for 12–72 h before electrophysiological measurements were made.

A cell line of HEK-293 cells stably transfected with human TREK-1 was maintained in 5% CO2 in a humidified incubator at 37°C in growth media (89.3% Dulbecco's modified Eagle's medium; 8.9% heat-inactivated foetal bovine serum; 1.8% geneticin (20 mg ml−1)). When the cells were 80% confluent, they were split and plated onto glass coverslips coated with poly-D-lysine (1 mg ml−1) and stored in an incubator at 5% CO2. The cells were used after 1–2 days for electrophysiological measurements. HEK-293 cells stably expressing human TREK-1 two-pore channels and the human TREK-1 (Meadows et al., 2000) and TASK-3 (Meadows & Randall, 2001) two-pore channel clones in the pcDNA 3.1 vector were kindly provided by Dr Helen Meadows at GlaxoSmithKline, U.K.

Mutations and truncations

To generate mutations and truncations, point mutations were introduced into the TREK-1 channel clones using the Quikchange kit (Stratagene, Amsterdam, The Netherlands). A pair of short (25–35 bases) complementary oligonucleotide primers, incorporating the intended mutation, was synthesised (MWG-Biotech, Ebersberg, Germany). To aid identification of successful mutants, a diagnostic enzyme restriction site was included in the primer sequence. Mutant DNA constructs were sequenced (MWG-Biotech) to confirm the introduction of the correct mutated bases.


The composition of the control extracellular solution was (in mM) 145 NaCl, 2.5 KCl, 3 MgCl2, 1 CaCl2 and 10 HEPES, (titrated to pH 7.4 with NaOH). Glass microelectrodes were pulled from thick-walled borosilicate glass capillaries using a two-stage vertical puller (Narishige). Fire-polished pipettes were back-filled with 0.2-μm-filtered intracellular solution (composition in mM: 150 KCl, 3 MgCl2, 5 EGTA, 10 HEPES (titrated to pH 7.4 with KOH)). Voltage-clamp recordings were made using either outside-out patches or the whole-cell recording technique. Cells and patches were usually held at −80 mV, and, either 1 s voltage ramps from −120 to +60 mV or incrementing voltage steps from either −50 mV to +40 mV or −120 to +120 mV were recorded every 6 s. For the solutions above, the calculated potassium equilibrium potential was −103.4 mV. Therefore, only whole-cell recordings with positive holding current at −80 mV (thus where TREK-1 current dominates the resting conductance and there is a negligible leak conductance) were included in the analysis. Cells also had to have an access resistance <20 MΩ for inclusion.

Unlike whole-cell recording, outside-out patches from tsA 201 cells always contained a measurable ohmic leak conductance (see Gruss et al., 2004a). Current–voltage curves from outside-out patches were corrected for this ohmic leak using the equation Icorr=IobsIrevVm/Vrev where Icorr is the corrected current, Iobs is the observed current at a membrane potential Vm and Irev is the current measured at the K+ reversal potential Vrev.

Recordings were usually digitised at 2 kHz and the output of the patch-clamp amplifier (Axopatch 1D; Axon Instruments, Union City, CA, U.S.A.) was usually filtered at 1 kHz using pClamp software. All recordings were stored on a computer hard disk. All electrophysiological measurements were carried out at room temperature (21–23°C).

Drugs were applied by bath perfusion with complete exchange of solution within 90 s. Control TREK-1 currents showed some run down with time during recordings, although this varied from cell to cell. We minimised the influence of this rundown on the measured effect of drugs, by calculating the percentage inhibition by fluoxetine or norfluoxetine relative to the average of controls taken just prior and just after the drug application. Values throughout this paper are given as mean±standard error of the mean (s.e.m.). Statistical significance was assessed using the Student's t-test.


Fluoxetine inhibition of hTREK-1 channel currents

Whole-cell currents through TREK-1 channels were recorded in tsA 201 cells transiently transfected with the 2-PK channel TREK-1. Ramp changes in membrane potential were applied from −120 to +60 mV. A typical control current obtained from such a ramp is shown in Figure 1a. The current is outwardly rectifying with a reversal potential at around −100 mV, close to the calculated K equilibrium potential for the external and internal recording solutions used here (see Methods).

Figure 1
Fluoxetine inhibits TREK-1 channel currents. (a) TREK-1 currents evoked by ramp changes in voltage in control conditions and in the presence of increasing concentrations of fluoxetine. (b) Time-course plot of fluoxetine inhibition of TREK-1 current. Each ...

Bath application of fluoxetine (1–100 μM) produced a concentration-dependent inhibition of TREK-1 currents (Figure 1a and b), which was easily reversible on wash. The concentration–response curve (Figure 1c) gave an IC50 for inhibition of 19±2 μM (mean±s.e.m.) with a Hill coefficient of 0.9±0.1 and a maximum inhibition, obtained from a free fit of the data, close to 100%. TREK-1 channels stably transfected into HEK-293 cells had very similar sensitivity to fluoxetine, with an IC50 of 14±5 μM.

The voltage-dependence of block by fluoxetine was calculated by comparing a control ramp trace with one in the presence of 30 μM fluoxetine. The percentage inhibition at each voltage on the ramp protocol was calculated and plotted against the appropriate voltage (Figure 1d). The inhibition does not change significantly in the voltage range −60 to +60 mV, suggesting that block by fluoxetine is not voltage-dependent.

Although TREK-1 is a leak potassium conductance, it does show some voltage-dependence, the degree of which depends on the phosphorylation state of the channel (Bockenhauer et al., 2001). This is most clearly illustrated using step rather than ramp voltage protocols (Figure 2a), where a time-dependent component to the current is seen at test potentials of 0 mV and above. With step changes in voltage, the whole-cell TREK-1 current at +60 mV can be divided into an ‘instantaneous' and a ‘time-dependent' current (the former measured 0.5 ms after the step to +60 mV, while the total current was measured at the end of the test step). In such recordings, the instantaneous current at +60 mV was only 0.58±0.08 (n=8) of the total current. Fluoxetine (10 μM) was equally effective at inhibiting the instantaneous component of current (47±8%, n=8) and the total (54±6%, n=8) current (Figure 2a and b).

Figure 2
Fluoxetine is equally effective at blocking the instantaneous and sustained components of TREK-1 current. TREK-1 currents evoked by step changes in voltage from −120 to +120 mV in the absence (a) and presence (b) of 10 ...

Fluoxetine inhibition of TASK-3 current

It is of interest that fluoxetine was much less effective at inhibiting a related human 2-PK channel from a different subfamily, TASK-3, compared with its action on TREK-1 channels. Fluoxetine (100 μM) produced a 31.4±5.5% inhibition of these TASK-3 channels (n=6; Figure 3). This was significantly smaller (P<0.01) than block of TREK-1 channels, where 100 μM fluoxetine inhibited current by 84.1±3.3% (n=10). Thus, fluoxetine is more potent at blocking TREK-1 channels compared to TASK-3 channels.

Figure 3
Fluoxetine is much less effective at inhibiting TASK-3 channel currents compared to TREK-1. (a) Time-course plot of fluoxetine inhibition of TASK-3 current. Each point is a 5 ms average of the current at 0 mV taken from the ramp voltage ...

Norfluoxetine inhibition of hTREK-1 channel currents

The effect of norfluoxetine (the active metabolite of fluoxetine) on TREK-1 channels was also investigated. Like fluoxetine, norfluoxetine produced a concentration-dependent block of the channel (Figure 4). The concentration response curve for norfluoxetine showed that the metabolite was even more potent than the parent fluoxetine molecule (Figure 4c) with an IC50 for inhibition of 9±1 μM, a Hill coefficient of 1.3±0.1 and, again, a maximum inhibition close to 100%. As for fluoxetine, block by norfluoxetine was voltage-independent (Figure 4d).

Figure 4
Norfluoxetine inhibits TREK-1 channel currents. (a) TREK-1 currents evoked by ramp changes in voltage in control conditions and in the presence of increasing concentrations of norfluoxetine. (b) Time-course plot of norfluoxetine inhibition of TREK-1 current. ...

TREK-1 mutant channel properties

Many of the compounds that are known to interfere with TREK-1 channel activity (such as arachidonic acid, anaesthetic agents and phosphorylation by PKA) exert at least part of their action on the C-terminus of the channel (e.g. Patel et al., 1999). Similarly, at least part of the transduction of these effects relies on the amino acid E (glutamate) at position 306 (Honore et al., 2002). We have created a number of mutated TREK-1 channels and investigated whether these mutations have any effect on inhibition by fluoxetine. In doing so, we found that the mutations often affected the properties of the channels themselves.

In our hands, truncation of the last 89 amino acids (Δ89) of the C-terminus of TREK-1 resulted in a large reduction in channel function (Figure 5a–c). Conversely, making the mutation E306A resulted in a significant increase of the basal current measured at 0 mV from 98.2±12.3 pA/pF for wild-type channels to 189.7±39.7 pA/pF (Student's t-test, P=0.03) (see also Honore et al., 2002).

Figure 5
TREK-1 current density and channel kinetics are altered by channel mutations. (a) TREK-1 current density (pA/pF) at 0 mV for wild-type TREK-1 channels and mutated TREK-1 channels. (b and c) Average current density versus voltage plots ...

As well as increasing current density, the E306A mutation resulted in the loss of the slower phase of TREK-1 activation kinetics (Figure 5d and e). The instantaneous current measured at 0 mV was 0.86±0.02 of the total current for wild-type channels. This was significantly smaller (P<0.01) than that seen for E306A-mutated channels where the instantaneous current was 1.00±0.01 of the total current (Figure 5f).

The channel activity of Δ89 TREK-1 could be recovered in a number of ways. Acidification of the intracellular solution together with the use of outside-out macropatches (to ensure pH change close to the intracellular face of the channels) recovered channel activity (Figure 6b). Similarly, the channel activity of Δ89 could be recovered by introducing the E306A mutation to the truncated channel (see Figure 5c). Similar to the E306A mutation by itself, a significant (P<0.01) increase in the basal current to 197.7±35.8 pA/pF was observed.

Figure 6
Fluoxetine inhibition of TREK-1 is unaffected by C-terminus truncations but reduced for E306A-mutated channels. (a) Whole-cell currents obtained by ramp changes in voltage for Δ48 TREK-1 channels in the presence and absence of fluoxetine (100 ...

Fluoxetine inhibition of C-terminal-truncated TREK-1 channels

The C-terminus of TREK-1 is critical for the action of a number of compounds that regulate this channel (e.g. Patel et al., 1998; Gruss et al., 2004a). However, progressive deletion of the C-terminus did not affect fluoxetine inhibition of TREK-1. Unlike the longer truncation, Δ48 displays normal channel function. Fluoxetine (100 μM) produced a block of 94.9% (Figure 6a) comparable with that of wild-type TREK-1 channels.

After recovery of Δ89 TREK-1 channel function, using internal acidification to pH 5 in outside out patches, inhibition by fluoxetine was found to be 85.8±2.8% (Figure 6b). This was not significantly different from wild type, which is inhibited by 84.1±3.3%.

Fluoxetine inhibition of E306A TREK-1 channels

Mutation of amino acid E (glutamate) to A (alanine) has a profound effect on the function and regulation of TREK-1 channels (Honore et al., 2002). Similarly, we found that the magnitude of inhibition of TREK-1 by fluoxetine was significantly reduced by the E306A mutation with 100 μM fluoxetine producing a 39.5±4.9 % inhibition (P<0.01) (Figure 6c and d).

Interestingly, the double mutant Δ89(E306A) also exhibited a significant (P<0.01) reduction in the inhibition by fluoxetine (100 μM) to 60.8±4.7%. This suggests that the E306A mutation dominates the properties of the double mutant channels both in terms of current amplitude and degree of inhibition by fluoxetine.


We have shown that the TREK-1 2-PK channel can be inhibited, potently, by fluoxetine, with an IC50 of 19 μM. The active metabolite of fluoxetine, norfluoxetine, was even more potent than the parent compound with an IC50 of 9 μM.

We have found that the TREK-1 channel is considerably more sensitive to block by fluoxetine than a related 2-PK channel from a different subfamily, TASK-3. It is interesting to note that a recent paper by Hajdu et al. (2003) found that fluoxetine also does not have a potent inhibitory action on another 2-PK channel, TASK-1. As we have found for TASK-3, 100 μM concentrations were needed to produce a significant inhibition of TASK-1. Thus, it appears that fluoxetine is a more potent inhibitor of TREK-1 channels, at least compared to members of the TASK subfamily of 2-PK channels.

Functional model of TREK-1 channel gating

TREK-1 channels exhibit two functional states. Firstly, they can behave as a leak conductance open across the physiological voltage range. In this state, the channels inhibit depolarisation and therefore decrease the excitability of the cell. Secondly, the channels can exist in a voltage-dependent state that preferentially passes an outward current at depolarised potentials. Thus, in this outward rectifier state, TREK-1 channels do not impede the depolarisation towards action potential firing threshold, but facilitate recovery and repetitive firing (Bockenhauer et al., 2001). The existence of these two states allows for a fine-tuning in the regulation of neuronal excitability. It has been shown that TREK-1 channels can be reversibly converted between their two phenotypes, via phosphorylation by protein kinase A (Bockenhauer et al., 2001). It is proposed that when the serine residue S333 in the C-terminal domain is phosphorylated, the channel becomes voltage-dependent. In normal, physiological conditions, the channel population exists as a mixture of phosphorylated and dephosphorylated channels and, following depolarising voltage-steps, two components to the activation of the current are observed – an instantaneous and a time-dependent component.

Another important residue close to the C-terminus of TREK-1 is E306, which has been proposed to act as an intracellular proton sensor. It has been shown that the mutation of the glutamate (E) to an alanine (A) at this position produces a channel that is a constitutively active leak conductance, mimicking the effects of intracellular acidosis (Honore et al., 2002, see also results). E306A can also rescue both lipid- and mechanosensitivity of a loss of function TREK-1 mutant (Honore et al., 2002).

Taking this information together, Honore et al. (2002) have proposed a kinetic model of TREK-1 channel gating containing two closed states (C1 and C2) and one open state (O) (see Figure 7). In the proposed model, the channel moves from C1 to C2 via internal acidosis or dephosphorylation of S333. Conversely, the channel can move from C2 to C1 by phosphorylation of S333. The instantaneous component of the whole-cell current following a voltage-step thus represents channels in C2 moving to O, while the time-dependent component represents channels moving more slowly to O from C1 (see also Honore et al., 2002). Since both the instantaneous and time-dependent components of the current were blocked equally well by fluoxetine, we propose that fluoxetine (and norfluoxetine) can bind to TREK-1 channels equally well, whether they are in C1 or C2 (Figure 7). Thus, we propose that fluoxetine acts as an allosteric blocker of TREK-1 channels binding to the channel in its closed states rather than acting as an open-channel blocking agent (see Hille, 2001 and below). A similar model has been proposed for a number of drugs that block potassium channels, such as ketoconazole block of KV1.5 channels (Iftinca et al., 2001).

Figure 7
Model for TREK-1 gating (adapted from Honore et al., 2002).

The mutation E306A is thought to mimic acidosis, and thus move the channel to C2. Without the ability to return to C1, the amount of time spent by the channel in O should be increased. In accordance with this, an increase in the basal current of E306A TREK-1 compared to the wild type was observed. This conversion of TREK-1 to a constitutively active channel demonstrates that the E306 residue is critical for gating of the channel. It was also observed that the activation kinetics of E306A TREK-1 were altered. Instead of the two components of activation seen in the wild type, only the instantaneous component was observed. Thus, the E306A mutation ‘locks' the channel in the dephosphorylated, voltage-independent state (C2). The inhibition of E306A by fluoxetine was shown to have the same time course and (lack of) voltage-dependence of inhibition as the wild type, but the magnitude of the inhibition was significantly reduced. In this mutation, fluoxetine can only bind to C2, from which the channel has a higher probability of returning to O. A reduced inhibition of E306A by fluoxetine results from the higher probability of the channel escaping from C2 (or drug bound C2) to O.

The truncation of the C-terminus of TREK-1 (Δ89) resulted, in our hands, in a loss of function channel. We speculate that the channel may be ‘stuck' in C1 and unable to reach O. Application of internal acidosis promotes the transition to C2, and restores the equilibrium between the two closed states. Channel function and inhibition by fluoxetine are therefore restored as observed. It is also possible to recover the truncated channel function by introducing the E306A mutation. This is proposed to work in the same manner as internal acidosis. However, it is evident that the Δ89(E306A) mutant of TREK-1 is unable to restore wild-type behaviour precisely, as the inhibition by fluoxetine is lower than in the wild type.

Site of action of fluoxetine

There is mounting evidence that fluoxetine, like other antidepressant agents (see Mathie et al., 1998), can inhibit a variety of neuronal ion channels including voltage-gated Ca2+ channels (Deak et al., 2000), Na2+ channels (Pancrazio et al., 1998) and various KV channels (Choi et al., 1999; 2001; Yeung et al., 1999). For all of these channels, an exact mechanism of block has yet to be determined, but there has been some evidence to suggest that in KV1.3 channels, fluoxetine acts at an intracellular site to cause an open-channel block (Choi et al., 1999). In addition, it has been shown that norfluoxetine inhibits KV1.3 channels with a higher potency than fluoxetine (Choi et al., 1999). A similar channel-blocking mechanism has been proposed for the inhibition of KV3.1 by norfluoxetine (Choi et al., 2001) and, indeed, for the inhibition of HERG K channels by fluoxetine (Thomas et al., 2002). This is not, however, a universal mechanism. For example, Yeung et al. (1999) showed that block of native KV channels in cerebellar granule cells by fluoxetine was neither voltage- nor use-dependent.

For TREK-1 channels, block by fluoxetine is not voltage-dependent. Furthermore, fluoxetine is less effective at blocking the E306A-mutated TREK-1 channels despite the channels having a higher open probability (see above and Honore et al., 2002). This suggests that the blocking site for fluoxetine in TREK-1 channels is unlikely to be in the channel pore itself and argues against an open-channel block mechanism for fluoxetine in this case.

The block of TREK-1 by fluoxetine also does not seem to involve the C-terminus of the channel. Truncation of the C-terminus by 48 or 89 amino acids, while interfering with channel function, did not appear to influence the degree of channel block by the drug. This can be contrasted with the effects of volatile anaesthetic agents such as chloroform and halothane (Patel et al., 1999), and gaseous anaesthetic agents such as cyclopropane (Gruss et al., 2004a). Augmentation of TREK-1 currents by these agents was either completely abolished (Patel et al., 1999) or considerably reduced (Gruss et al., 2004a) by progressive deletion of the C-terminus of TREK-1.

While our data suggest that fluoxetine does not bind to either the pore region of TREK-1 channels or the C-terminus, it is not clear, at present, where fluoxetine does bind. Being a lipophilic compound, one intriguing possibility is that the mechanism of fluoxetine block may be analogous to that of antipsychotic agent chlorpromazine, which has been suggested to block TREK-1 channels by entering the lipid membrane and acting as a cationic amphipathic cup-former (Patel et al., 1998), thereby changing the curvature of the lipid bilayer. In this regard, it is of interest that both the high-affinity block of the voltage-dependent K channel (KvAP) and the block of a stretch-activated ion channel by two different toxins (VSTX1) and (GsMTx4) isolated from the venom of the tarantula Grammostola spatulata have been attributed to the ability of these toxins to partition into the lipid membrane (Lee & MacKinnon, 2004; Suchyna et al., 2004), a phenomenon termed ‘mechanopharmacology' (Suchyna et al., 2004).

Possible clinical consequences of fluoxetine block of TREK-1 channels

The therapeutic plasma concentration of fluoxetine is estimated as between 0.15 and 1.5 μM (see Orsulak et al., 1988; Altamura et al., 1994). Under steady-state conditions, the plasma concentrations of norfluoxetine are higher than those of fluoxetine (Hiemke & Hartter, 2000) and, for both compounds, can reach much higher levels in the brain during chronic fluoxetine treatment (Karson et al., 1993). Fluoxetine is preferentially distributed from plasma into brain tissue because of its high lipophilicity (Karson et al., 1993). In addition, both compounds have long half-lives of elimination (4–6 days for fluoxetine and 4–16 days for norfluoxetine), so these compounds persist in the body for several weeks following discontinuation of treatment. Recent fluorine magnetic resonance spectroscopy measurements of fluoxetine have shown that the free concentration throughout the whole brain is around 5 μM for adults on an average dose of 24 mg per day (Strauss et al., 2002). It is worth noting that many patients take considerably higher doses than this; the recommended daily dose for major depressive episodes and obsessive compulsive disorder is 20–60 mg per day, while for bulimia nervosa it is 60 mg per day (www.emc.medicines.org.uk; summary of product characteristics (SPC) for ProzacR). Thus, the steady-state concentrations of both fluoxetine and norfluoxetine in the brain are sufficient to produce significant block of TREK-1 channels.

Fluoxetine is thought to exert its antidepressant effects by selectively inhibiting the reuptake of 5-hydroxytryptamine (5-HT, serotonin) from the synaptic cleft (Wong et al., 1995). Fluoxetine inhibits 5-HT uptake in synaptosomal preparations of rat brain with a Ki of 8–20 nM, while norfluoxetine has a Ki of 17–45 nM (Wong et al., 1995). It is possible, however, that block of TREK-1 channels may contribute to the antidepressant action of this drug. Antidepressant drugs have been reported to decrease hypothalamic temperature and it has been suggested that this function, during sleep, may contribute to the antidepressant action of at least some of these agents (Duncan et al., 1995). Since TREK-1 is a heat-activated channel, which is highly expressed in the hypothalamus (Maingret et al., 2000a), it is possible that fluoxetine block of TREK-1 contributes to its antidepressant action. In addition, there are moderate levels of TREK-1 expression in the serotonergic, raphe nuclei, but little expression in the noradrenergic, locus ceruleus (Talley et al., 2001; Gu et al., 2002).

More likely, however, is that block of TREK-1 channels by fluoxetine and norfluoxetine may contribute to a number of the known side effects of fluoxetine treatment. Recent experiments on TREK-1 knockout mice have demonstrated the importance of these channels in neuroprotection against epilepsy and ischaemia, and in underlying at least part of the action of certain anaesthetic agents such as the volatile anaesthetics (Heurteaux et al., 2004). Since fluoxetine blocks TREK-1 channels, this may exacerbate any tendency to convulsions and may underlie the warning that fluoxetine should be ‘introduced cautiously' in patients who have a history of seizures and avoided in patients with unstable seizure disorders/epilepsy (www.emc.medicines.org.uk; SPC for ProzacR). This presents a particular management problem for clinicians, since epileptic patients often also suffer from depression (Salzberg & Vajda, 2001).

In conclusion, this study has shown that fluoxetine and norfluoxetine are inhibitors of the 2-PK channel TREK-1 and may block this channel in treated patients. Our results obtained from both wild-type and mutant TREK-1 channels are consistent with the functional model of TREK-1 gating proposed by Honore et al. (2002).


This work was supported by the MRC. LEK is supported by an MRC Studentship and SJA was supported by a Wellcome Trust Vacation Scholarship. We thank Raquel Yustos for technical assistance and Helen Meadows (Glaxo SmithKline) for the gift of TREK-1 and TASK-3 clones.


two-pore domain potassium
protein kinase A
standard error of the mean
summary of product characteristics


  • ALEXANDER S.P.H., MATHIE A., PETERS J.A. Potassium channels, in Guide to receptors and channels. Br. J. Pharmacol. 2004;141:S83–S84.
  • ALTAMURA A.C., MORO A.R., PERCUDANI M. Clinical pharmacokinetics of fluoxetine. Clin. Pharmacokinet. 1994;26:201–214. [PubMed]
  • BLONDEAU N., LAURITZEN I., WIDMANN C., LAZDUNSKI M., HEURTEAUX C. A potent protective role of lysophospholipids against global cerebral ischemia and glutamate excitotoxicity in neuronal cultures. J. Cereb. Blood Flow Metab. 2002;22:821–834. [PubMed]
  • BOCKENHAUER D., ZILBERBERG N., GOLDSTEIN S.A. KCNK2: reversible conversion of a hippocampal potassium leak into a voltage-dependent channel. Nat. Neurosci. 2001;4:486–491. [PubMed]
  • CHEMIN J., GIRARD C., DUPRAT F., LESAGE F., ROMEY G., LAZDUNSKI M. Mechanisms underlying excitatory effects of group I metabotropic glutamate receptors via inhibition of 2P domain K+ channels. EMBO J. 2003;22:1–9. [PMC free article] [PubMed]
  • CHOI B.H., CHOI J.S., YOON S.H., RHIE D.J., MIN D.S., JO Y.H., KIM M.S., HAHN S.J. Effects of norfluoxetine, the major metabolite of fluoxetine, on the cloned neuronal potassium channel Kv3.1. Neuropharmacology. 2001;41:443–453. [PubMed]
  • CHOI J.S., HAHN S.J., RHIE D.J., YOON S.H., JO Y.H., KIM M.S. Mechanism of fluoxetine block of cloned voltage-activated potassium channel Kv1.3. J. Pharmacol. Exp. Ther. 1999;291:1–6. [PubMed]
  • CHUMBLEY J.R., RANATUNGA K.M., MATHIE A. Inhibition of the human two-pore domain K+ channel TREK-1 (hTREK-1) by fluoxetine. Br. J. Pharmacol. 2003;138:183P.
  • DEAK F., LASZTOCZI B., PACHER P., PETHEO G.L., KECSKEMETI V., SPAT A. Inhibition of voltage-gated calcium channels by fluoxetine in rat hippocampal pyramidal cells. Neuropharmacology. 2000;39:1029–1036. [PubMed]
  • DUNCAN W.C., JOHNSON K.A., WEHR T.A. Antidepressant drug-induced hypothalamic cooling in Syrian hamsters. Neuropsychopharmacology. 1995;1:17–37. [PubMed]
  • FINK M., DUPRAT F., LESAGE F., REYES R., HEURTEAUX C., LAZDUNSKI M. Cloning, functional expression and brain localisation of a novel unconventional outward rectifier K+ channel. EMBO J. 1996;15:6854–6862. [PMC free article] [PubMed]
  • GOLDSTEIN S.A., BOCKENHAUER D., O'KELLY I., ZILBERBERG N. Potassium leak channels and the KCNK family of two-P-domain subunits. Nat. Rev. Neurosci. 2001;2:175–184. [PubMed]
  • GRUSS M., BUSHELL T.J., BRIGHT D.P., LIEB W.R., MATHIE A., FRANKS N.P. Two-pore-domain K+ channels are a novel target for the anesthetic gases xenon, nitrous oxide and cyclopropane. Mol. Pharmacol. 2004a;65:443–452. [PubMed]
  • GRUSS M., MATHIE A., LIEB W.R., FRANKS N.P. The two-pore-domain K+ channels TREK-1 and TASK-3 are differentially modulated by copper and zinc. Mol. Pharmacol. 2004b;66:530–537. [PubMed]
  • GU W., SCHLICHTHORL G., HIRSCH J.R., ENGELS H., KARSCHIN C., KARSCHIN A., DERST C., STEINLEIN O.K., DAUT J. Expression pattern and functional characteristics of two novel splice variants of the two-pore-domain potassium channel TREK-2. J. Physiol. 2002;539:657–668. [PMC free article] [PubMed]
  • HAJDU P., ULENS C., PANYI G., TYTGAT J. Drug- and mutagenesis-induced changes in the selectivity filter of a cardiac two-pore background K+ channel. Cardiovasc. Res. 2003;58:46–54. [PubMed]
  • HERVIEU G.J., CLUDERAY J.E., GRAY C.W., GREEN P.J., RANSON J.L., RANDALL A.D., MEADOWS H.J. Distribution and expression of TREK-1, a two pore domain potassium channel in adult rat CNS. Neuroscience. 2001;103:899–919. [PubMed]
  • HEURTEAUX C., GUY N., LAIGLE C., BLONDEAU N., DUPRAT F., MAZZUCA M., LANG-LAZDUNSKI L., WIDMANN C., ZANZOURI M., ROMEY G., LAZDUNSKI M. TREK-1, a K+ channel involved in neuroprotection and general anaesthesia. EMBO J. 2004;23:2684–2695. [PMC free article] [PubMed]
  • HIEMKE C., HARTTER S. Pharmacokinetics of selective serotonin reuptake inhibitors. Pharmacol. Ther. 2000;85:11–28. [PubMed]
  • HILLE B. Ionic Channels of Excitable Membranes. Sunderland: Sinauer Associates Inc; 2001.
  • HONORE E., MAINGRET F., LAZDUNSKI M., PATEL A.J. An intracellular proton sensor commands lipid- and mechano-gating of the K+ channel TREK-1. EMBO J. 2002;21:2968–2976. [PMC free article] [PubMed]
  • IFTINCA M., WALDRON G.J., TRIGGLE C.R., COLE W.C. State-dependent block of rabbit vascular smooth muscle delayed rectifier and Kv1.5 channels by inhibitors of cytochrome P450-dependent enzymes. J. Pharmacol. Exp. Ther. 2001;298:718–728. [PubMed]
  • KARSON C.N., NEWTON J.E., LIVINGSTON R., JOLLY J.B., COOPER T.B., SPRIGG J., KOMOROWSKI R.A. Human brain fluoxetine concentrations. J. Neuropsychiatry Clin. Neurosci. 1993;5:322–329. [PubMed]
  • KENNARD L.E., VEALE E.L., MATHIE A. Inhibition of the human two-pore domain potassium channel, TREK1, by fluoxetine and its metabolite norfluoxetine. FENS Abstr. 2004;2:A082.9. [PMC free article] [PubMed]
  • KOH S.D., MONAGHAN K., SERGEANT G.P., RO S., WALKER R.L., SANDERS K.M., HOROWITZ B. TREK-1 regulation by nitric oxide and cGMP-dependent protein kinase. An essential role in smooth muscle inhibitory neurotransmission. J. Biol. Chem. 2003;276:44338–44346. [PubMed]
  • LAURITZEN I., BLONDEAU N., HEURTEAUX C., WIDMANN C., ROMEY G., LAZDUNSKI M. Poly-unsaturated fatty acids are potent neuroprotectors. EMBO J. 2000;19:1784–1793. [PMC free article] [PubMed]
  • LEE S.-Y., MACKINNON R. A membrane-access mechanism of ion channel inhibition by voltage sensor toxins from spider venom. Nature. 2004;430:232–235. [PubMed]
  • LESAGE F. Pharmacology of neuronal background potassium channels. Neuropharmacology. 2003;44:1–7. [PubMed]
  • MAINGRET F., LAURITZEN I., PATEL A., HEURTEAUX C., REYES R., LESAGE F., LAZDUNSKI M., HONORE E. TREK-1 is a heat-activated background K+ channel. EMBO J. 2000a;19:2483–2491. [PMC free article] [PubMed]
  • MAINGRET F., PATEL A.J., LESAGE F., LAZDUNSKI M., HONORE E. Mechano- or acid stimulation, two interactive modes of activation of the TREK-1 potassium channel. J. Biol. Chem. 1999;274:26691–26696. [PubMed]
  • MAINGRET F., PATEL A.J., LESAGE F., LAZDUNSKI M., HONORE E. Lysophospholipids open the two P domain mechano-gated K+ channels TREK-1 and TRAAK. J. Biol. Chem. 2000b;275:10128–10133. [PubMed]
  • MATHIE A., WOOLTORTON J.R.A., WATKINS C.S. Voltage-activated potassium channels in mammalian neurons and their block by novel pharmacological agents. Gen. Pharmacol. 1998;30:13–24. [PubMed]
  • MEADOWS H.J., BENHAM C.D., CAIRNS W., GLOGER I., JENNINGS C., MEDHURST A.D., MURDOCK P., CHAPMAN C.G. Cloning, localisation and functional expression of the human orthologue of the TREK-1 potassium channel. Pflugers Arch. 2000;439:714–722. [PubMed]
  • MEADOWS H.J., RANDALL A.D. Functional characterization of human TASK-3, an acid-sensitive two-pore domain potassium channel. Neuropharmacology. 2001;40:551–559. [PubMed]
  • O'CONNELL A.D., MORTON M.J., HUNTER M. Two-pore domain K+ channels – molecular sensors. Biochem. Biophys. Acta. 2002;1566:152–161. [PubMed]
  • ORSULAK P.J., KENNEY J.T., DEBUS J.R., CROWLEY G., WITTMAN P.D. Determination of the antidepressant fluoxetine and its metabolite norfluoxetine in serum by reversed-phase HPLC with ultraviolet detection. Clin. Chem. 1988;34:1875–1878. [PubMed]
  • PANCRAZIO J.J., KAMATCHI G.L., ROSCOE A.K., LYNCH C., III Inhibition of neuronal Na+ channels by antidepressant drugs. J. Pharmacol. Exp. Ther. 1998;284:208–214. [PubMed]
  • PATEL A.J., HONORE E., LESAGE F., FINK M., ROMEY G., LAZDUNSKI M. Inhalational anesthetics activate two-pore-domain background K+ channels. Nat. Neurosci. 1999;2:422–426. [PubMed]
  • PATEL A.J., HONORE E., MAINGRET F., LESAGE F., FINK M., DUPRAT F., LAZDUNSKI M. A mammalian two pore domain mechano-gated S-like K+ channel. EMBO J. 1998;17:4283–4290. [PMC free article] [PubMed]
  • SALZBERG M., VAJDA F.J.E. Epilepsy, depression and antidepressant drugs. J. Clin. Neurosci. 2001;8:209–215. [PubMed]
  • STRAUSS W.L., UNIS A.S., COWAN C., DAWSON G., DAGER S.R. Fluorine magnetic resonance spectroscopy measurement of brain fluvoxamine and fluoxetine in pediatric patients treated for pervasive developmental disorders. Am. J. Psychiatry. 2002;159:755–760. [PubMed]
  • SUCHYNA T.M., TAPE S.E., KOEPPE R.E., ANDERSEN O.S., SACHS F., GOTTLIEB P.A. Bilayer-dependent inhibition of mechanosensitive channels by neuroactive peptide enantiomers. Nature. 2004;430:235–240. [PubMed]
  • TALLEY E.M., SOLORZANO G., LEI Q., KIM D., BAYLISS D.A. CNS distribution of members of the two-pore-domain (KCNK) potassium channel family. J. Neurosci. 2001;21:7491–7505. [PubMed]
  • THOMAS D., GUT B., WENDT-NORDAHL G., KIEHN J. The antidepressant drug fluoxetine is an inhibitor of human ether-a-go-go-related gene (HERG) potassium channels. J. Pharmacol. Exp. Ther. 2002;300:543–548. [PubMed]
  • WONG D.T., BYMASTER F.P., ENGLEMAN E.A. Prozac (fluoxetine, Lilly 110140), the first selective serotonin uptake inhibitor and an antidepressant drug: twenty years since its first publication. Life Sci. 1995;57:411–441. [PubMed]
  • YEUNG S.Y., MILLAR J.A., MATHIE A. Inhibition of neuronal KV potassium currents by the antidepressant drug, fluoxetine. Br. J. Pharmacol. 1999;128:1609–1615. [PMC free article] [PubMed]

Articles from British Journal of Pharmacology are provided here courtesy of The British Pharmacological Society
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...