Abstract
Inhalation anesthetics activate and cannabinoid agonists inhibit TWIK-related acid-sensitive K+ channels (TASK)-1 two-pore domain leak K+ channels in vitro. Many neuromodulators, such as noradrenaline, might also manifest some of their actions by modifying TASK channel activity. Here, we have characterized the basal behavioral phenotype of TASK-1 knockout mice and tested their sensitivity to the inhalation anesthetics halothane and isoflurane, the α2 adrenoreceptor agonist dexmedetomidine, and the cannabinoid agonist WIN55212-2 mesylate [R-(+)-[2,3-dihydro-5-methyl-3-[(morpholinyl)methyl]pyrrolo[1,2,3,-de]-1,4-benzoxazinyl]-(1-naphtalenyl)methanone mesylate)]. TASK-1 knockout mice had a largely normal behavioral phenotype. Male, but not female, knockout mice displayed an enhanced acoustic startle response. The knockout mice showed increased sensitivity to thermal nociception in a hot-plate test but not in a tail-flick test. The analgesic, sedative, and hypothermic effects of WIN55212-2 (2–6 mg/kg s.c.) were reduced in TASK-1 knockout mice. These results implicate TASK-1-containing channels in supraspinal pain pathways, in particular those modulated by endogenous cannabinoids. TASK-1 knockout mice were less sensitive to the anesthetic effects of halothane and isoflurane than wild-type littermates, requiring higher anesthetic concentrations to induce immobility as reflected by loss of the tail-withdrawal reflex. Our results support the idea that the activation of multiple background K+ channels is crucial for the high potency of inhalation anesthetics. Furthermore, TASK-1 knockout mice were less sensitive to the sedative effects of dexmedetomidine (0.03 mg/kg s.c.), suggesting a role for the TASK-1 channels in the modulation of function of the adrenergic locus coeruleus nuclei and/or other neuronal systems.
The TASK-1 (K2P3.1, KCNK3) and -3 (K2P9.1, KCNK9) K+ channel subunits are members of the two-pore domain background K+ (K2P) channel family (Duprat et al., 1997; Leonoudakis et al., 1998; Kim et al., 2000; Lopes et al., 2000). K2P channels contribute to the resting membrane potential and action potential duration (see Goldstein et al., 2001; Patel and Honoré, 2001; Bayliss et al., 2003). The K+ flow-through TASK-1 and -3 channels are outward and relatively independent of voltage and time. Inhibition of these channels promotes membrane depolarization, whereas their activation induces hyperpolarization. Both TASK-1 and -3 channels are sensitive to small pH changes within the physiological range [inhibition by acidic conditions, pK 7.4 for TASK-1, hence their name TWIK-related acid-sensitive K+ channels (Duprat et al., 1997)]. TASK-1 and -3 can each form functional homomeric channels (subunit dimers), but they can also form heterodimers in cell types where the two genes are both expressed (Berg et al., 2004; Aller et al., 2005).
TASK-1/-3 currents are regulated by neuromodulators (e.g., noradrenaline, serotonin, H+, and Zn2+) (Talley et al., 2000; Chemin et al., 2003; Meuth et al., 2003; Aller et al., 2005). Often, the neuromodulators act via G protein-coupled receptors to promote phosphatidylinositol bisphosphate depletion from the TASK-1/-3 channels, therefore causing channel closure and membrane depolarization (Chemin et al., 2003; Meuth et al., 2003). Some agonist drugs acting on these modulatory receptors have clinical applications; e.g. dexmedetomidine, an α2-adrenergic receptor agonist, is used as a sedative and analgesic in intensive care units and in veterinary practice. High densities of α2-adrenoreceptors and TASK-1/-3 channels occur in the locus coeruleus (Talley et al., 2001), the source of forebrain noradrenergic innervation. The hypnotic and sedative effects of α2-adrenoceptor activation arise by inhibiting the locus coeruleus, but it is not clear whether TASK-1/-3 channels are involved directly (or indirectly) with the actions of dexmedetomidine.
Considering further the neuromodulatory theme, recombinant TASK-1 and -3 channels are blocked by the endocannabinoid anandamide and the synthetic cannabinoid agonist WIN55212-2 (Maingret et al., 2001; Berg et al., 2004). Therefore, the inhibition of TASK-1 channels could be an important pharmacological mechanism of anandamide, because some of its in vivo effects are not mediated via the central nervous system CB1 cannabinoid receptor (Adams et al., 1998; Di Marzo et al., 2000; Monory et al., 2002).
Altering K2P activity via G proteins is used by neuromodulators to alter the dynamic balance of circuit activity, but strongly opening these channels by direct agonist gating has been hypothesized to induce unconsciousness, i.e., deep anesthesia. Inhalation anesthetics directly and potently activate several K2P channels, including TASK-1 and -3, in transfected cells, arterial chemoreceptors, and rat brain slices (Patel et al., 1999; Sirois et al., 2000; Talley and Bayliss, 2002; Washburn et al., 2002; Berg et al., 2004; for review, see Franks and Honore, 2004). Indeed, genetic deletion of the K2P channel TREK-1 (K2P2.1, KCNK2) reduced, but not abolished, the sensitivity of mice to inhalation anesthetics, confirming the in vivo contribution of K2P channel activation to the actions of inhalation anesthetics (Heurteaux et al., 2004). The residual sensitivity to inhalation anesthetics might indicate the contribution of multiple K2P channel types or other targets such as ligand-gated ion channel receptors (for review, see Sonner et al., 2003; Franks and Honoré, 2004; Rudolph and Antkowiak, 2004).
Here, we analyzed the basal behavioral phenotype of adult TASK-1 knockout mice and tested their sensitivity to the inhalation anesthetics halothane and isoflurane as well as the cannabinoid agonist WIN55212-2 and the α2 adrenoceptor agonist dexmedetomidine. We wished to explore to what extent the actions of these various modulators and anesthetics depended on their ability to activate (or inhibit) TASK-1-containing channels.
Materials and Methods
TASK-1 Knockout Mice. TASK-1 knockout mice were generated as described previously with the genetic background, C57BL/6J ×129S1/SvJ (Aller et al., 2005). Homozygous knockout and wild-type littermates used in the present study were from the fifth and sixth generations of heterozygous breeding.
Mice were bred and genotyped in Heidelberg. For behavioral studies, they were transferred to Helsinki, where they were allowed to equilibrate in their new environment for 3 weeks. Adult male and female TASK-1 knockout and littermate wild-type mice were maintained in the standard animal facilities in groups of one to five in polypropylene macrolon cages with food pellets and tap water available ad libitum. Lights were on from 6:00 AM to 6:00 PM. All animal tests were approved by the Laboratory Animal Committee of the University of Helsinki and the Southern Finland Provincial Government. Basal behavioral characterization was performed before pharmacological studies. When the same mice were used for several tests, at least a 1-week washout period was kept between experiments.
In Situ Hybridization of TASK-1 and -3 in Wild-Type Mouse Brain. In situ hybridization with 35S-labeled TASK-1- and -3-specific oligonucleotide probes was performed as described earlier (Wisden and Morris, 1994; Aller et al., 2005). Brains were from male wild-type adult mice (n = 3, C57BL/6J). To increase the hybridization signal for TASK-1, two different TASK-1 oligonucleotides were hybridized together. Both oligonucleotide probes for TASK-1 gave identical results (Aller et al., 2005).
Real-Time Quantitative RT-PCR of K2P Channels. Using Ultraspec RNA reagents (Biotecx, Houston, TX), total RNA from forebrain was isolated from adult TASK-1 knockout (n = 4) and wild-type littermate (n = 4) mice. RNAs were treated for 15 min with DNase I (QIAGEN, Valencia, CA) and cleaned with an RNeasy Mini kit (QIAGEN). Reverse transcription was performed with 5 μg of total RNA using a SuperScript Double-Stranded cDNA synthesis kit (Invitrogen, San Diego, CA). Real-time PCR was performed with the following protocol: 10 min at 95°C, 15 s at 95°C, and 1 min at 60°C (40 cycles) using a SYBR Green PCR kit (PE Applied Biosystems, Diagnostik, Weiterstadt, Germany) and a Gene Amp 5700 sequence detector (PE Biosystems). A linear concentration-amplification curve was established by diluting pooled samples. Quantified results for individual cDNAs were normalized to cyclophilin. Each experiment was performed in triplicate. The primers were as described by Aller et al. (2005).
Basal Behavioral Characterization. Behavioral and physiological characterization of phenotypes was performed as described by Vekovischeva et al. (2004). The person who observed and recorded the behavior was not aware of the genotype of the tested animals. The mice were evaluated by observing them for 45 s in a viewing jar for body position, defecation, urination, spontaneous activity, respiratory rate, and presence of tremor. After transferring the mouse in the open arena, transfer arousal, locomotor activity, gait, pelvic and tail elevations, and piloerection were evaluated during a 30-s observation period. Then grip strength, lacrimation, salivation, provoked biting, negative geotaxis reflex, and righting reflex, as well as startle, pinna, corneal, and toe-pinch reflexes, were evaluated. In addition, touch escape, struggle behavior, and vocalization during handling were evaluated.
To measure the stress-induced hyperthermia, the mice were transferred to the testing room at least 1 h before the experiment. The mice were familiar with human handling, because this experiment was performed after elevated plus-maze and staircase tests. Basal rectal temperature was recorded (BAT-12, Physitemp, Clifton, NJ), and a mouse was placed immediately in an empty 500-ml glass jar for 10 min, after which the rectal temperature was measured again. Mice were not returned to their home cage until all mice from the same cage had been analyzed.
To study the anxiety phenotype of TASK-1 knockout mice, their behavior on an elevated plus-maze was analyzed after SHIRPA screening so that mice were practically naive to handling except standard animal care. The elevated plus-maze apparatus was made of gray plastic and elevated to 50 cm from the floor level. It consisted of a central platform (5 × 5 cm), from which two open arms (5 × 40 cm with a 0.7-cm margin) and two closed arms (5 × 40 × 20 cm) extended. The mice were placed individually on the central platform facing an open arm, and the experimenter immediately left the testing room. The mice were allowed free exploration of the maze for 5 min, and their behavior was recorded using a video tracking system with a charge-coupled device video camera above the maze. The position and movements of the center of the animal's surface area were analyzed automatically using EthoVision Color-Pro 3.0 software (Noldus Information Technology, Wageningen, The Netherlands). The central area was extended to include the first 2 cm of each arm. An arm entry was recorded when the center of the mouse entered the arm area defined as the distal 38 cm of the arm. This corresponds to the definition of an arm entry when all four legs are on the arm. The maze was cleaned with water-moistened paper towel after each mouse. The mice were returned to their home cage when all mice from the same cage had been tested.
To measure basal locomotor activity and exploratory behavior, the mice were placed individually in a staircase box, which was made of gray plastic and consisted of five stairs (2 cm high, 10 cm wide, and 7.5 cm deep) surrounded by a 12-cm-high wall. Behavior was recorded using a charge-coupled device video camera above the staircase. Behavior was observed from a video monitor in an adjacent room and scored manually (number of rears and stairs climbed up) or automatically (total distance moved, latency to leave the lowest platform for the first time) using EthoVision. For the EthoVision analysis, the area of the lowest platform was extended to include proximal 3 cm of the first stair so that an entry to the stairs was recorded when the center of the mouse left the extended platform area, corresponding the definition of all four legs on the stair. The staircase was cleaned with water-moistened paper towel after each mouse. Locomotor activity was also studied in the staircase 30 min after vehicle or WIN55212-2 (6 mg/kg s.c.) administration.
Acoustic Startle Response and Prepulse Inhibition. The acoustic startle response and prepulse inhibition (PPI) were measured using a two-unit automated startle system (Med Associates Startle Reflex System; Med Associates Inc., Georgia, VT). In an illuminated and sound-attenuated chamber, a small cage (7.5 cm long, 3.5 cm wide, and 4.0 cm high) with metal bars was mounted above a piezoelectric sensor. Movements in the cylinder were detected by the sensor, digitized, and analyzed by Startle Reflex System software (version 4.01; Med Associates Inc.). Sensitivity of two chambers were calibrated and adjusted to be identical (Platform calibrator; Med Associates Inc.). A 65-dB background noise and acoustic stimuli were delivered through speakers in a ceiling of the chambers. The study design was based on recently described mouse procedures (Heldt et al., 2004). Mice were acclimated for the cylinders and chambers for 5 min daily for 3 days before the experiments. Then, acoustic startle responses to stimuli of different intensities were determined. A test session began with a 5-min acclimation. Background noise (65 dB) was on during the acclimation period and throughout the session. The acclimation period was followed by seven blocks of trials containing seven stimuli (40 ms) of different intensities (65, 71, 75, 85, 95, 110, and 120 dB) in a pseudorandom order. Intertrial interval varied between 9 and 21 s. The test session contained 49 trials and lasted approximately 20 min.
The prepulse inhibition was analyzed 1 week later. First, the mice were acclimated to the chambers for 5 min with the 65-dB background noise on. To habituate the mice to stimuli, the test session began with six trials of 95- or 110-dB stimuli (40 ms), which were not included in the data analysis. This was followed by 10 blocks of four trials containing both 95- and 110-dB startle stimuli alone and combined to a 71-dB prepulse stimulus (20 ms). The prepulse stimulus was delivered also alone five times during the session. The order of different type of stimuli was pseudorandom. The interval between prepulse and startle stimuli was 100 ms, and the intertrial interval varied between 9 and 21 s. The test session contained 51 trials and lasted approximately 20 min.
Movements of the mouse were sampled 50 ms before the first stimulus (null period) and 200 ms after the first stimulus. Startle amplitude was defined as peak amplitude that occurred during the first 100 ms after the onset of the startle stimulus. If the movements of the animal produced a clear signal (peak value ± 150) during a null period, the trial was excluded from the results. The exclusion was done by an experimenter viewing graphs from each trial blind to the type of stimulus, animal gender, and genotype. The percent PPI was calculated from the formula: %PPI = [(amplitude of startle pulse alone–amplitude of startle pulse when preceded by prepulse)/amplitude of startle pulse alone] × 100.
Hot-Plate and Tail-Flick Tests. To avoid stress-induced analgesia, the mice were habituated to the hot-plate test apparatus (Hot Plate Analgesic Meter; Harvard Apparatus, Edenbridge, UK) for 5 days by placing a mouse on this plate at room temperature inside a plastic cylinder (20 cm in diameter) daily for 3 min. During the test, the surface was maintained at 52 ± 0.2°C. The latency to react was scored when the mouse rapidly moved, licked its hind paw, or jumped. The cut-off time was 40 s. Before drug or vehicle administration, two basal reaction times, 15 min apart, were determined for each mouse. Then, the latencies to react were analyzed 30 min after vehicle, WIN55212-2 (2 mg/kg s.c.), or morphine (3 and 6 mg/kg s.c.) injections.
The mice were acclimated to the tail-flick test procedure and apparatus (Model DS20; Ugo Basile, Comerio, Italy) for 7 days by gently holding them immobile inside a cloth for 3 min. The heat intensity was adjusted so that the basal reaction time was 2 to 5 s in C57BL/6J mice (intensity of light ∼14.5 V). The cut-off time was 7 s. On the distal part of the tail, a 1-cm area was marked with a black pen, and heat was directed to this area. Before drug or vehicle administration, two basal reaction times, 15 min apart, were determined. Then, the latencies to the withdrawal reaction were analyzed 30 min after vehicle and WIN55212-2 (4 mg/kg s.c.) injections. The analgesic responses were calculated as percentage maximal possible effect [% analgesia = (test latency–basal latency)/(cut-off time–basal latency) × 100].
Sensitivity to Inhalation Anesthetics. Halothane anesthesia was induced and maintained in a 5-liter chamber that was warmed (33–34°C) by a heat radiator placed above to maintain the normal body temperature during anesthesia (rectal temperatures after the experiments ranged from 35.9–38.7°C; there were no differences between the genotypes). Halothane was added to the air stream (4 l/min) by a vaporizer (Fluotec 3; Ohmeda, BOC Health Care, Westyorkshire, UK). Carbon dioxide and halothane concentrations were monitored by Capnomax-Ultima (Datex, Instrumentarium, Helsinki, Finland). Carbon dioxide concentration was maintained at <0.3% by placing soda lime granules onto the floor of the chamber. During the 20-min equilibrium time, the righting reflex was analyzed every 2 min by turning the mouse in a supine position consecutively three times. The mouse was considered to have lost its righting reflex (LORR) when it was not able to right itself in every three consecutive trials, and the concentration of the anesthetic was recorded. After the LORR, the mouse was left in the chamber at the same concentration until the end of the 20-min equilibrium time when the tail-clamp withdrawal reflex was analyzed at this concentration. Thereafter, the tail-clamp withdrawal reflex was tested following 20-min equilibrium times at higher anesthetic concentration levels. Animals were considered to have lost the tail-clamp withdrawal reflex (LOTW) when they did not respond by movement of an extremity or jerking of the head to the clamp of the proximal part of the tail for 5 s with a hemostat covered with plastic tubing to prevent skin damage. Halothane concentrations used were 0.5, 0.7, 0.9, 1.1, 1.3, and 1.5%. Isoflurane anesthesia (Isotec 5 vaporizer; Ohmeda, BOC Health Care) was induced and maintained in a similar way as halothane anesthesia, except that the air stream was faster (7 l/min) and equilibrium time was shorter (6 min). Isoflurane concentrations used were 0.6, 0.9, 1.2, 1.5, and 1.8%.
Dexmedetomidine does not induce LORR in mice, even at high doses (Paris et al., 2003). Therefore, its sedative effect was determined by recording the locomotor activity of wild-type and TASK-1 knockout mice in an open arena. The mouse was placed individually in an empty cage (55 × 33 × 18 cm) with a white plastic floor marked with 15 squares (11 × 11 cm), and its behavior was recorded for 3 min using a video camera. The latency to leave the center square, the total number of squares entered, and the total number of rears were analyzed 30 min after vehicle or dexmedetomidine (0.03 mg/kg s.c.) administration. The mouse was returned to its home cage, and the rectal temperature was measured 60 min after vehicle or dexmedetomidine administration.
Drugs. WIN55212-2 mesylate was obtained from Tocris (Bristol, UK) and dissolved first in 100% cremophor (Sigma, St. Louis, MO) and then diluted with saline to a final concentration of 0.2 to 0.6 mg/ml (5–10% cremophor). Morphine hydrochloride (20 mg/ml injection) solution was obtained from Leiras (Turku, Finland) and diluted with saline to a final concentration of 0.3 mg/ml. Halothane was obtained from Rhodia (Bristol, UK), and isoflurane was from Abbott Laboratories (Queensborough, Kent, UK). Dexmedetomidine hydrochloride (0.1 mg/ml injection solution; Orion Pharma, Espoo, Finland) was diluted with water to a final concentration of 0.006 mg/ml. Pregnanolone (5β-pregnan-3α-ol-20-one) was obtained from Sigma.
WIN55212-2-Stimulated Guanosine-5′-O-(3-[35S]thio)-Triphosphate Autoradiography. WIN55212-2-stimulated GTPγ[35S] autoradiography was performed as described earlier with minor modifications (Sim-Selley and Martin, 2002). Adult TASK-1 knockout (n = 6) and wild-type littermate (n = 6) mice were decapitated, and whole brains were frozen in isopentane on dry ice and stored at –80°C. Horizontal 14-μm-thick serial sections were cut with a cryostat (Leica CM3050 S; Leica Microsystems, Nussloch, Germany) and thaw-mounted on SuperFrost object glasses. The sections were preincubated in assay buffer (50 mM Tris-HCl, pH 7.4, 5 mM MgCl, 100 mM NaCl, 1 mM EDTA, and 0.5% bovine serum albumin) for 20 min at 20°C. The sections were transferred to the assay buffer containing 2 mM GDP and 1 μM 8-cyclopentyl-1,3-dipropylxanthine (DPCPX; Sigma) and incubated for 60 min at 20°C. DPCPX was added to eliminate basal adenosine A1 receptor-dependent G protein activity. The sections were then incubated in the absence (basal) or with 0.01, 0.1, 1, 10, or 33 μM WIN55212-2 in assay buffer containing 0.078 nM GTPγ[35S] (specific activity, 1250 Ci/mmol; PerkinElmer Life and Analytical Sciences, Boston, MA), 1.3 mM dithiothreitol, 2 mM GDP, and 1 μM DPCPX for 90 min at 20°C. Nonspecific binding was determined by incubating the sections with 10 μM unlabeled GTP (Sigma). The sections were washed twice in ice-cold 50 mM Tris-HCl, pH 7.4, and 5 mM MgCl for 5 min, dipped in ice-cold deionized water, and dried in airflow at room temperature. The sections were exposed to Kodak Biomax MR film (Eastman Kodak, Rochester, NY) for 1 week with 14C radioactivity standards (GE Healthcare, Little Chalfont, Buckinghamshire, UK). Binding densities in selected brain areas were quantified with MCID M5 imaging software (Imaging Research, St. Catharines, ON, Canada) and converted to nanocuries per gram of radioactivity values on the basis of the simultaneously exposed standards.
Electrophysiological Recordings of Recombinant TASK-1 and -3 Channels. Modified HEK-293 cells (tsA-201) were incubated in growth medium (89% Dulbecco's modified Eagle's medium, 10% heat-inactivated fetal bovine serum, 10,000 units/ml 1% penicillin, and 10 mg/ml streptomycin) under 5% CO2 at 37°C. When the tsA-201 cells were 80% confluent, they were split and plated for transfection onto glass coverslips coated with poly-d-lysine (1 mg/ml) to ensure good cell adhesion. The tsA-201 cells were transiently transfected using the calcium phosphate method. One microgram of cDNA expression vector encoding a mouse or human TASK-1 or -3 subunit was added to each 15-mm well, and 1 μg of a plasmid encoding the cDNA of green fluorescent protein was included to identify cells expressing K2P channels. After 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 to 60 h before electrophysiological measurements were made. Human TASK-1 and -3 (Meadows and Randall, 2001) K2P channel clones in the pcDNA 3.1 vector were kindly provided by Dr. Helen Meadows (GlaxoSmithKline, Uxbridge, Middlesex, UK). Mouse TASK-1 and -3 expression vectors were as described by Aller et al. (2005).
Whole-cell currents were recorded from tsA-201 cells transiently transfected with mTASK-3, hTASK-3, mTASK-1, or hTASK-1 channels. The composition of the control extracellular solution was 145 mM NaCl, 2.5 mM KCl, 3 mM MgCl2, 1 mM CaCl2, and 10 mM HEPES (titrated to pH 7.4 with NaOH). Microelectrodes were pulled from thick-walled borosilicate glass capillaries. Fire-polished pipettes were back-filled with 0.2-μm-filtered intracellular solution (150 mM KCl, 3 mM MgCl2, 5 mM EGTA, and 10 mM HEPES titrated to pH 7.4 with KOH). Voltage-clamp recordings were made using the whole-cell recording technique. Cells were held at –80 mV and then subjected to a step to –40 mV for 500 ms, followed by a 500-ms voltage-ramp from –110 to + 20 mV, once every 5 s. Recordings were digitized at 10 kHz, and the record was filtered at 5 kHz. All electrophysiological measurements were carried out at room temperature (21–23°C).
Statistical Testing. Two-way analysis of variance (ANOVA) was used to test main gender and genotype effects (SPSS 10.0.7 for Windows; SPSS Inc., Chicago, IL). If no significant gender effect was found, the data from males and females were combined and tested further using one-way ANOVA followed by Newman-Keuls or Dunnett's post hoc tests or using Student's t test or Mann-Whitney analyses (Prism 3.03; GraphPad Software, Inc., San Diego, CA). Data from motor training and acoustic startle experiments were tested using repeated measures two-way ANOVA followed by one-way ANOVA and Newman-Keuls or Dunnett's post hoc tests. A limit for significance was p < 0.05.
Results
Coexpression and Divergent Expression of the TASK-1 and -3 Genes in the Adult Mouse Brain. To identify potential brain areas (in addition to cerebellar granule cells; Aller et al., 2005) whose function might be compromised by deleting TASK-1 expression, we hybridized a series of coronal brain sections (adult wild-type mouse) from the olfactory bulb to the hind brain with TASK-1- and -3-specific oligonucleotide probes (Fig. 1). The images shown in Fig. 1 have the same exposure time for TASK-1 and -3, but the TASK-1 sections were hybridized with two different TASK-1 oligonucleotide probes simultaneously to increase the hybridization signal.
At the level of X-ray films, the brain areas with the highest TASK-1 gene expression are the cerebellar granule cells (Fig. 1M), motor neurons (for example, motor nucleus 7, Fig. 1M), pontine nuclei, interpeduncular nucleus (Fig. 1I), locus coeruleus (Fig. 1M), and medial mammillary nucleus (Fig. 1G), followed by the olfactory bulb granule cells (Fig. 1A). There is some detectable TASK-1 gene expression in the neocortex, raphe nuclei, reticular thalamus (Fig. 1E), certain central thalamic nuclei (Fig. 1E), parafascicular thalamic nucleus (Fig. 1G), and medial geniculate nucleus (Fig. 1, I and K). The expression patterns of TASK-1 and -3 clearly differ in some thalamic, hypothalamic, and pontine nuclei.
The mouse brain areas with both TASK-1 and -3 transcripts include the cerebellar granule cells, locus coeruleus, motor neurons (Fig. 1, M and N), pontine nuclei (although higher in TASK-1) (Fig. 1, I and J), and possibly some cells in the neocortex (TASK-1 is uniformly weak in neocortex, TASK-3 expression is more prominent), habenula, olfactory bulb granule cells, and cells in the external plexiform layer of the olfactory bulb (Fig. 1, A and B). In these areas, TASK-1/TASK-3 heterodimers may be replaced by homomeric TASK-3 channels in TASK-1 knockout brains.
Unchanged mRNA Levels of Other K2P Channels in the TASK-1 Knockout Forebrain. Real-time quantitative RT-PCR was employed to examine possible changes in K2P expression as a result of silencing the TASK-1 gene. We selected the most highly expressed K2P channels present in the adult forebrain: TASK-1 and -3, TREK-1 and -2, TWIK-1, TRAAK, and THIK-2. Apart from the loss of TASK-1 mRNA in the TASK-1 knockout mice, we could detect no significant difference in the steady-state transcript levels of the other K2P gene family members in the forebrain of TASK-1 knockout and wild-type littermate mice (Fig. 2). This is consistent with our earlier findings that the mRNA expression levels of the other K2P gene family members were unaltered in the cerebellum of TASK-1 knockout mice (Aller et al., 2005). In addition, in situ hybridization signals for the mRNAs of other K2P gene family members were similar in brain horizontal sections from wild-type and TASK-1 knockout mice (Aller et al., 2005).
TASK-1 Knockout Mice Display a Largely Normal Behavioral Phenotype. Two cohorts (fifth and sixth generations) of adult TASK-1 knockout and littermate wild-type mice, both males and females, were tested using the SHIRPA primary screening protocol (Table 1). It revealed a grossly normal behavioral phenotype for TASK-1 knockout mice. The only difference was in a toe-pinch response, which was affected by both gender and genotype (Table 1). The response was significantly (p < 0.05, Mann-Whitney) weaker in knockout females compared with wild-type females.
The anxiety-related behavior in the elevated plus-maze test, such as percentages of the time spent on open arms, was not affected by TASK-1 deletion (Table 1). Likewise, no significant difference was observed in stress-induced hyperthermia between wild-type and knockout mice (Table 1). Moreover, no difference was observed in the locomotor activity (total distance moved during 5 min) on the elevated plusmaze between wild-type and knockout mice. In a staircase test that measures locomotor and exploratory activity, no significant difference was observed in most parameters (Table 1). However, in the latency to leave the lowest platform, there was a significant genotype effect (F1,64 = 6.83, p < 0.05) but no gender effect. The knockout mice entered other stairs of the staircase significantly (p < 0.05, Student's t test) faster than the wild-type mice.
Enhanced Acoustic Startle Response in Male TASK-1 Knockout Mice. We examined the sensorimotor functions of TASK-1 knockout mice by studying the acoustic startle response and prepulse inhibition of the response. The pulse intensity (65–120 dB) significantly (F2,228 = 54.04, p < 0.001) affected the acoustic startle responses. There was also a significant gender-genotype interaction (F2,228 = 9.69, p < 0.01); therefore, the data of males and females were analyzed separately (Fig. 3). In males, the startle amplitudes were affected by pulse intensity (F1,117 = 36.60, p < 0.001), the effect being significant at 110- and 120-dB intensities (p < 0.01, Dunnett's post hoc test) (Fig. 3A). The knockout males responded with significantly (F1,117 = 12.62, p < 0.01) stronger startles than wild-type littermates to the pulses of 110 and 120 dB (p < 0.001, Newman-Keuls post hoc test) (Fig. 3A). In females, a significant effect of pulse intensity (F1,110 = 20.27, p < 0.001) was found, being significant at 110- and 120-dB intensities (p < 0.01, Dunnett's post hoc test) (Fig. 3B). No genotype effect was observed among the females.
To estimate a threshold signal that generates a significant startle response compared with a baseline response (65 dB signal), paired comparisons using Student's t test were separately performed within each group. Mean responses to 65-dB signal were compared with mean responses to signals of varying decibels of the same animals. Both male and female knockout mice responded to 75-dB signal with significantly stronger amplitude compared with their baseline responses. The threshold signals for wild-type males were 85 dB, and those for wild-type females were 95 dB.
Next, we studied whether the TASK-1 deletion affected prepulse inhibition. Because the startle amplitudes of wild-type males were lower than those of knockout males, different pulse intensities were used for wild-type and knockout males (110 and 95 dB, respectively). For females, the pulse intensity was similar (95 dB). The same 71-dB prepulse intensity was used for all groups. At these pulse intensities, the startle amplitudes were similar for all groups (Fig. 3C). No statistically significant difference was observed between wild-type and knockout male and female mice in the prepulse inhibition (Fig. 3D). A trend to a smaller prepulse inhibition in female wild-type mice compared with male wild-type mice is consistent with the reported gender differences in humans and weaker prepulse inhibition in women (Swerdlow et al., 1993).
TASK-1/-3 Channels Are Insensitive to Pregnanolone. Given that TASK-1 knockout male and female mice differed strongly in their responses to acoustic stimuli, we speculated that, in some neuronal pathways of TASK-1 knockout mice, TASK-3 homodimers might have replaced TASK-1/TASK-3 heterodimers and that these TASK-3 channels could differ in their steroid sensitivity, therefore possibly explaining the gender difference. To our knowledge, steroid sensitivity has not been studied on recombinant TASK channels. Thus, we transfected tsA-201 cells with expression vectors for TASK-1 and -3 (human and mouse) and tested the sensitivity of the resulting leak K+ currents to pregnanolone (5β-pregnan-3α-ol-20-one). There was no effect or very little effect of pregnanolone on either channel type. For hTASK-3 currents, we observed 2 ± 3% (n = 3) inhibition at 100 nM and 1 ± 4% (n = 4) inhibition at 1 μM pregnanolone. For mTASK-3 currents, we observed a very slight 8 ± 2% (n = 6) inhibition by 1 μM pregnanolone. For both hTASK-1 and mTASK-1 currents, we saw no effect of pregnanolone (1 μM) with 0 ± 7% (n = 3) and 2 ± 5% (n = 5) inhibition, respectively.
Enhanced Sensitivity to Thermal Nociception and Reduced Analgesic Effect of WIN55212-2 in TASK-1 Knockout Mice. On the basis of the small defects observed in TASK-1 knockout mice in hind paw withdrawal response to toe-pinch stimulus (Table 1) and the expression of TASK-1 currents in the spinal cord and dorsal root ganglia (Talley et al., 2001; Cooper et al., 2004), we next examined responses of the knockout mice to thermal nociception. In the hot-plate test, there was a significant genotype effect (F1,31 = 8.79, p < 0.01; Fig. 4A). No significant gender effect was observed in two-way ANOVA, and when the data from males and females were combined, the knockout mice had a significantly (p < 0.05, Student's t test) shorter latency to react by licking or shaking a hind paw or by jumping than wild-type littermates when placed on the hot plate.
Cannabinoid agonists anandamide, methanandamide, and WIN55212-2 inhibit TASK channels (Maingret et al., 2001; Berg et al., 2004; Aller et al., 2005). Cannabinoid agonists also have well established analgesic effects, and endogenous cannabinoids may modulate nociceptive pathways (Jaggar et al., 1998). Here, we tested whether the analgesic effect of WIN55212-2 has been affected by the inactivation of TASK-1 in knockout mice. WIN55212-2 (2 mg/kg s.c.) produced a significant antinociception (F2,63 = 43.36, p < 0.001) in the hot-plate test (Fig. 4C). A significant drug-genotype interaction was observed (F2,63 = 5.99, p < 0.05). In addition, the gender and genotype effects were significant (F2,63 = 6.46, p < 0.05; F2,63 = 4.47, p < 0.05, respectively). When the data from males were separately tested, a significant drug effect (F1,32 = 44.51, p < 0.001) was observed; the effect of WIN55212-2 compared with vehicle was significant in both wild-type and knockout male mice in Newman-Keuls post hoc tests (Fig. 4C). In addition, there was a significant drug-genotype interaction (F1,32 = 5.56, p < 0.05), which was apparently due to a significantly (p < 0.05, Newman-Keuls) smaller effect of WIN55212-2 in TASK-1 knockout mice compared with wild-type mice (Fig. 4C). When the data from females were separately tested, no interaction or genotype effects were found, but a significant drug effect (F1,30 = 8.83, p < 0.05) was observed. The analgesic effect of WIN55212-2 was significant only in wild-type mice (Fig. 4C).
In the other thermal nociception test, the tail-flick test, wild-type and knockout male and female mice had no difference in the latency to withdrawal reflex after exposure to thermal stimulus (Fig. 4B). With drug treatment, significant drug (F2,59 = 36.10, p < 0.001) and gender (F2,59 = 6.29, p < 0.05) effects, but no genotype effects (Fig. 4D), were observed. WIN55212-2 (4 mg/kg s.c.) increased the analgesic response both in males (F1,32 = 37.79, p < 0.001) and females (F1,28 = 8.19, p < 0.05), but only in males, the post hoc tests also revealed a statistically significant increase in percentage analgesia (Fig. 4D).
The analgesic effect of morphine (3 mg/kg and cumulative 6 mg/kg s.c.) was also compared between wild-type and TASK-1 knockout mice in the hot-plate test. The basal response to thermal nociception differed between wild-type and knockout mice (genotype effect, F1,31 = 11.18, p < 0.01; data not shown) similarly as in the earlier experiment (Fig. 4A). To be able to use two doses of morphine within the same experiment, the effect of vehicle was not determined because the analgesia produced by vehicle was small and did not differ between genotypes or genders in the earlier WIN55212-2 experiment (Fig. 4C). Two-way ANOVA did not reveal any gender or genotype effect at either dose of morphine (Fig. 4E).
Reduced Sensitivity to Hypomotility and Hypothermic Effects of WIN55212-2 in TASK-1 Knockout Mice. The effect of WIN55212-2 (6 mg/kg s.c.) was tested on the locomotor activity and body temperature of TASK-1 knockout mice to investigate whether WIN55212-2 actions on TASK-1 had effects additional to the supraspinal antinociceptive ones. First, we analyzed staircase behavior and rectal temperature after vehicle administration 1 day before WIN55212-2 administration. No significant difference between wild-type and knockout mice was observed in total distance moved, number of rears, or latency to leave the lowest platform (Fig. 5A). Likewise, the rectal temperature after vehicle injection was not different between the genotypes, although a significant gender effect (F1,30 = 18.14, p < 0.001; wild-type male 36.3 ± 0.3°C versus knockout male 36.8 ± 0.2°C and wild-type female 37.5 ± 0.1°C versus knockout female 37.4 ± 0.2°C) was observed as before (cf. Table 1). On the next day, the staircase behavior and rectal temperature were recorded 30 min after administration of WIN55212-2 (6 mg/kg s.c.). There were significant gender effects on all three staircase parameters measured (F1,30 = 6.34 for distance moved, F1,30 = 6.47 for number of rears, F1,30 = 7.98 for latency to leave the lowest platform) (Fig. 5B). Significant genotype effects were observed in the total distance moved (F1,30 = 4.89, p < 0.05) and the number of rears (F1,30 = 10.15, p < 0.01). A significant gender-genotype interaction was observed in the number of rears (F1,30 = 5.02, p < 0.05). The knockout males moved less after WIN55212-2 than the knockout females, whereas the locomotor activity of wild-type males and females was similarly reduced by WIN55212-2 (Fig. 5B). Importantly, the wild-type females were more sedated and exhibited less rearing after WIN55212-2 than the knockout females (Fig. 5B). The latency to move (time of complete immobility) after placement on the lowest platform in the staircase was also measured. The genotype affected the immobility times (F1,30 = 4.67, p < 0.05), but the post hoc analysis failed to support the shorter immobility times in the knockout animals (Fig. 5C). The hypothermia 60 min after WIN55212-2 (6 mg/kg s.c.) administration was significantly affected both by gender (F1,30 = 7.63, p < 0.05) and genotype (F1,30 = 7.42, p < 0.05), but post hoc comparison did not show significant differences between treatment groups (Fig. 5D).
Unaltered WIN55212-2 Stimulation of GTPγ[35S] Binding in TASK-1 Knockout Brain. Since behavioral responses to WIN55212-2 were attenuated in TASK-1 knockout mice, we investigated whether cannabinoid CB1 receptor-mediated effects are intact in the knockout brain. CB1 receptor-regulated G protein activation can be measured using a GTPγ[35S] binding assay that is sensitive enough to detect changes in CB1 receptor desensitization or down-regulation (Sim-Selley and Martin, 2002). No difference was observed in basal GTPγ[35S] binding on wild-type or TASK-1 knockout brain sections (Fig. 6, A and B). Concentrations lower than 10 μM stimulated G protein activation minimally and without difference between the genotypes (Fig. 6C). Higher WIN55212-2 concentrations, 10 and 33 μM, increased GTPγ[35S] binding in the forebrain and cerebellum with a similar magnitude of stimulation in both genotypes (p ≥ 0.08; Fig. 6C).
Reduced Sensitivity to Halothane and Isoflurane in TASK-1 Knockout Mice. Inhalation anesthetics halothane and isoflurane activate recombinant TASK-1 and -3 channels and TASK-like currents in brain slices (Patel et al., 1999; Sirois et al., 2000; Washburn et al., 2002; Berg et al., 2004). At higher concentrations, isoflurane slightly inhibits TASK-1 homodimeric channels (Berg et al., 2004). We examined the sensitivity of TASK-1 knockout mice to these anesthetics by determining the concentrations at which the mice lose their righting (LORR) and tail-withdrawal (LOTW) reflexes. The anesthetic concentration that is required to suppress the withdrawal reaction to tail-clamp in 50% of mice was also determined [ED50 = minimum alveolar concentration (MAC)]. The genotype, but not the gender, significantly affected the halothane concentration at which the mice lost their tail-withdrawal reflex (F1,30 = 7.57, p < 0.05). TASK-1 knockout mice needed a higher halothane concentration at LOTW (Fig. 7B, p < 0.01) than the wild types, but the lower concentration effect, the LORR (Fig. 7A), was produced at similar halothane concentrations. In the dose-response curve, the percentage of animals showing no movement during a tail-clamp test (i.e., LOTW) was fitted for halothane concentrations (Fig. 7C). A shift to the right in the dose-response curve indicates the reduced sensitivity in TASK-1 knockout mice (MAC = 1.32; 95% confidence interval, 1.30–1.35) compared with wild-type mice (MAC = 1.21; 95% confidence interval, 1.16–1.26).
In isoflurane sensitivity tests, the genotype, but not the gender, affected the isoflurane concentration at which the mice had the LORR (F1,33 = 4.71, p < 0.05), but no significant effect was observed in the LOTW. The TASK-1 knockout mice required a higher isoflurane concentration for LORR than the wild-type mice (p < 0.05, Student's t test; Fig. 7D). A modest shift to the right in the dose-response curve in the LOTW also suggests the reduced sensitivity to isoflurane in TASK-1 knockout mice (MAC = 1.36; 95% confidence interval, 1.32–1.39) compared with wild-type mice (MAC = 1.27; 95% confidence interval, 1.24–1.29) (Fig. 7F).
Reduced Sensitivity to Dexmedetomidine in TASK-1 Knockout Mice. TASK-1 and -3 channels are expressed at a high level in the rat and mouse noradrenergic locus coeruleus (Talley et al., 2001; this study; Fig. 1M). To analyze whether the TASK-1 deletion affects the functions of noradrenergic neurons, we tested the sensitivity of TASK-1 knockout mice to dexmedetomidine, which produces strong sedative/hypnotic and hypothermic effects by inhibiting noradrenaline release through activation of presynaptic α2-adrenergic receptors (Lähdesmäki et al., 2004).
The effect of dexmedetomidine (0.03 mg/kg s.c.) on locomotor activity was tested in an open arena. To determine the basal behavior of wild-type and TASK-1 knockout mice, the locomotor activity was determined at 30 min and the rectal temperature was determined at 60 min after vehicle administration 1 day before dexmedetomidine administration. Two-way ANOVA did not reveal significant effects of gender or genotype in the number of squares entered, the number of rears, in the latency to leave the center square or in the rectal temperatures. Therefore, the data from males and females after vehicle administration were combined (Fig. 8A). On the next day, we analyzed the locomotor activity 30 min and the rectal temperature 60 min after dexmedetomidine (0.03 mg/kg s.c.) administration. Two-way ANOVA revealed a significant genotype, but not gender, effect in the total number of squares entered (F1,29 = 11.19, p < 0.01), the total number of rears (F1,29 = 7.23, p < 0.05), and in the temperature change (F1,29 = 4.58, p < 0.05). TASK-1 knockout mice moved and reared significantly more than wild-type mice after dexmedetomidine administration (Fig. 8B). The horizontal locomotor activity (the total number of squares entered) was reduced by dexmedetomidine to 16 ± 3 and 37 ± 5% of the corresponding vehicle values in wild-type and TASK-1 knockout mice, respectively (p < 0.01, Student's t test, Fig. 8B), and after dexmedetomidine, none of the wild-type mice performed any rears, whereas eight of 15 TASK-1 knockout mice did rear at least once (Fig. 8B). These results indicate that TASK-1 knockout mice were less sensitive to the sedative action of dexmedetomidine, which apparently was reflected also in a decrease in temperature (Fig. 8C).
Discussion
We have found that the actions of inhalation anesthetics, the α2 adrenergic agonist dexmedetomidine, and the cannabinoid agonist WIN55212-2 have changed in TASK-1 knockout mice. Although TASK-1 knockout mice displayed a largely normal behavioral and physiological phenotype, alterations observed in their basal behavior revealed an involvement of TASK-1-containing channels in the auditory and nociceptive responses. We believe that the changed responses to inhalational anesthetics and cannabinoid ligands are likely to be direct effects because these reagents directly bind the TASK channels, but the reduced sensitivity to dexmedetomidine may rather be an indirect effect of altered activity of the locus coeruleus or some other brain region since α2 adrenoreceptors couple through Gi/o, and most TASK-like effects seem to be through Gq/11 (Talley et al., 2000; Chemin et al., 2003; Meuth et al., 2003).
In the TASK-1 knockout mice, a mixture of neuronal effects will be produced. Some areas will have lost their TASK channels entirely, whereas others will still have functioning TASK-3 channels. TASK-1 expression is accompanied with high-TASK-3 expression in most brain regions, but a comparison of TASK-1 and -3 gene expression patterns revealed a few regions (e.g., reticular and central relay nuclei of the thalamus; mammillary nucleus; pontine nuclei) where TASK-1 expression predominates and where its deletion may lead to reduced K2P currents. On the other hand, areas such as the locus coeruleus might be expected to assemble heteromeric TASK-1/TASK-3 channels (see Fig. 1M).
The Contribution of TASK-1-Containing Channels to Sensorimotor Coordination. The acoustic startle responses to different pulse intensities were dramatically stronger in the knockout than wild-type males, whereas females responded with intermediate levels showing no genotype difference. The inhibition of the startle response by a weak prepulse was similar in wild-type and knockout mice, indicating normal sensorimotor gating and auditory function. Therefore, it is unlikely that the smaller startle amplitudes of wild-type male mice were due to hearing problems. In search for an explanation for this gender difference among the TASK-1 knockout mice, we tested TASK-1 and -3 channels for their sensitivity to a neuroactive steroid, pregnanolone. Although pregnanolone failed to affect recombinant TASK-1 or -3 channel-mediated currents, tests with other steroids are required to rule out their direct actions on TASK channels. Gender differences are common in knockout mouse strains and can be due to indirect actions of sex hormones during development or in adult brain. For example, neuroprotective actions of estrogen may change the effects of gene deletions in female mice (Behl and Holsboer, 1999).
The TASK-1 knockout mice may be slightly more reactive in a novel environment compared with wild-type littermates, because they had a significantly shorter latency to start exploring the staircase, increased pain sensitivity, and tended to have enhanced stress-induced hyperthermia. This reactivity is probably not related to increased fear or anxiety because wild-type and knockout mice behaved similarly in the elevated plus-maze test of anxiety. Increased reactivity to sensory stimuli is in line with the increased acoustic startle observed in male knockout mice. Moreover, the reduced sensitivity to the sedative effects of dexmedetomidine in the knockouts suggests alterations in the noradrenergic system, which may underlie enhanced acoustic startle response. Indeed, it has been reported that dexmedetomidine attenuates acoustic startle responses by reducing firing of noradrenergic neurons (Lähdesmäki et al., 2004). However, it is also possible that the disruption of TASK-1 enhances neuronal excitability outside the noradrenergic system. Therefore, further studies with different pharmacological agents that modulate inhibitory and excitatory neurotransmission are required to elucidate the overall excitability of TASK-1 knockout brain.
The Contribution of TASK-1-Containing Channels to Nociception and to Actions of Cannabinoids. TASK-1 knockout mice had an increased sensitivity to thermal nociception in the hot-plate test but not in the tail-flick test. The tail-flick test measures spinal reflexes (Irwin et al., 1951); therefore, the unchanged responses in this test suggest that the role of TASK-containing channels is not crucial for spinal reflexes, although TASK-like currents occur in the dorsal root ganglia and spinal cord dorsal horn (Kindler et al., 2000; Talley et al., 2001; Cooper et al., 2004). The increased sensitivity in the hot-plate test could be due to the lack of TASK-1 or TASK-1/TASK-3 channels in the thalamus or alterations in ascending pain pathways. Alternatively, descending pain pathways originating for example from the locus coeruleus or raphe nuclei could be affected by the lack of normal TASK-1 function (Sirois et al., 2000; Talley et al., 2001; Washburn et al., 2002). The reduced sensitivity to dexmedetomidine suggests an altered, perhaps overactive, noradrenergic system in TASK-1 knockout mice. On the other hand, the reduced antinociception by WIN55212-2 suggests that the enhanced sensitivity to nociception may be due to compromised analgesic actions of endogenous cannabinoids, such as anandamide (Jaggar et al., 1998); this cannabinoid ligand inhibits TASK-1 and -3 channels in vitro (Maingret et al., 2001; Berg et al., 2004; Aller et al., 2005).
In addition to attenuated antinociception by WIN55212-2, TASK-1 knockout mice showed reduced sensitivity to the drug's hypomotility and hypothermic effects. TASK-1 deletion had only a partial effect on WIN55212-2, suggesting that most of WIN55212-2 actions arise at the CB1 cannabinoid receptor. Nevertheless, in vivo effects of cannabinoids cannot be totally blocked by CB1 receptor antagonists (Adams et al., 1998; Di Marzo et al., 2000; Monory et al., 2002). Indeed, in cerebellar granule cells that significantly express both TASK-1 and -3 subunits (Brickley et al., 2001; Talley et al., 2001; Aller et al., 2005), G proteins can be activated by anandamide and WIN55212-2 through a novel binding site in CB1 receptor knockout brains (Monory et al., 2002). Therefore, we tested whether G protein activation is altered in TASK-1 knockout mice. However, the activation of G proteins by WIN55212-2 as analyzed by stimulation of GTPγ[35S] binding was similar in wild-type and TASK-1 knockout brain.
The Contributions of TASK-1-Containing Channels to the in Vivo Actions of Anesthetics. TASK-1 knockout mice showed a reduced sensitivity to the inhalation anesthetics halothane and isoflurane, reflected as a rightward shift (reduced potency) in the concentration-response plots. Our results, together with the reduced sensitivity of TREK-1 knockout mice to inhalation anesthetics (Heurteaux et al., 2004), show that the combined activation of multiple K2P channels together with direct actions on ligand-gated ion channels, such as glycine, GABAA, and N-methyl-d-aspartate receptors, is crucial for the analgesic and immobility effects of inhalation anesthetics (Sonner et al., 2003; Franks and Honoré, 2004; Rudolph and Antkowiak, 2004). It is possible that the reduced sensitivity to halothane was partly due to altered activity of the locus coeruleus noradrenergic system. Halothane enhances TASK-like currents, leading to reduced excitability of locus coeruleus neurons, which may mediate some components of halothane anesthesia (Sirois et al., 2000). The reduction in halothane sensitivity of the TASK-1 knockouts is approximately the same as for the TREK-1 knockouts. The more modest reduction of isoflurane sensitivity in TASK-1 knockout mice is consistent with the small effects of isoflurane on rat TASK-1 channels in vitro (Berg et al., 2004). Isoflurane tends to activate, have no effect, or even inhibit rat TASK-1 function depending on its concentration (Berg et al., 2004). The slightly higher concentration of isoflurane that the TASK-1 knockout mice required to lose their righting reflex suggests that the TASK-1 channel activation may be needed at lower isoflurane concentrations during the early phase of anesthesia.
Possible Roles of Various Combinations of TASK-1 and Other K2P Channel Subunits. In most tests, the behavior and drug responses of the TASK-1 knockout mice were only slightly altered, suggesting that the TASK-1-containing K2P channels are not indispensable but may be efficiently compensated by other mechanisms. In our previous characterization of TASK-1 knockout mice, which focused on cerebellar granule cells, we showed that the mice had modest impairments in motor coordination or balance in Rotorod and walking beam tests (Aller et al., 2005). TASK-1-deficient granule cells still had a normal K+ leak conductance, but the modulation of this current by extracellular Zn2+ and H+ was dramatically altered (Aller et al., 2005). Specifically, the Zn2+ sensitivity of the leak current was only observed in TASK-1 knockout granule cells, demonstrating the assembly of TASK-3 homodimers in the absence of TASK-1 because TASK-3 homodimers are Zn2+-sensitive, whereas TASK-1 homodimers and TASK-1/TASK-3 heterodimers are Zn2+-insensitive (Clarke et al., 2004; Aller et al., 2005). Similar phenomena could have occurred in other cell types, so that in the TASK-1 knockout mice all TASK channels would be inhibited by Zn2+. This could lead to attenuated hyperpolarization in brain regions where Zn2+ is coreleased with GABA or glutamate (for review, see Clarke et al., 2004), perhaps explaining the irritable behavioral phenotype and blunted drug reactions by several different compounds.
In conclusion, we found that TASK-1 knockout mice were largely normal. However, the male knockout mice had enhanced startle responses to acoustic stimuli, and both male and female knockout mice had enhanced nociception in the hot-plate test. These results suggest that TASK-1 or TASK-1/TASK-3 channel function is important for normal sensory motor functioning. These processes may be regulated by endogenous cannabinoids or noradrenergic system. Finally, the reduced sensitivity to halothane and isoflurane indicates that TASK-1-containing channels mediate partially the actions of inhalation anesthetics.
Acknowledgments
We thank Irme Preugschat-Gumprecht (Heidelberg, Germany) and Aira Säisä (Helsinki, Finland) for excellent technical assistance. W.W. thanks H. Monyer for support and encouragement.
Footnotes
-
This work was supported by the German Research Council (Grant DFG WI 1951/1-2) (to W.W.), by Volkswagen Stiftung (Grant I/78 554) (to W.W.), by the Academy of Finland (to E.R.K.), and by the Sigrid Juselius Foundation (to E.R.K.).
-
doi:10.1124/jpet.105.098525.
-
ABBREVIATIONS: TASK, TWIK-related acid-sensitive K+ channel; TWIK, tandem P domain weak inwardly rectifying K+ channels; K2P, two-pore-domain background K+; WIN55212-2 mesylate, R-(+)-[2,3-dihydro-5-methyl-3-[(morpholinyl)methyl]pyrrolo[1,2,3,-de]-1,4-benzoxazinyl]-(1-naphtalenyl)methanone mesylate); PPI, prepulse inhibition; LORR, loss of righting reflex; LOTW, loss of tail withdrawal reflex; GTPγ[35S], guanosine-5′-O-(3-[35S]thio)-triphosphate; DPCPX, 8-cyclopentyl-1,3-dipropylxanthine; ANOVA, analysis of variance; MAC, minimum alveolar concentration; TREK, TWIK-related K+ channel.
-
↵1 Current affiliation: Institute of Medical Sciences, University of Aberdeen, Foresterhill, Aberdeen, Scotland.
- Received November 10, 2005.
- Accepted January 4, 2006.
- The American Society for Pharmacology and Experimental Therapeutics