KV7 Channels Regulate Firing during Synaptic Integration in GABAergic Striatal Neurons

Striatal projection neurons (SPNs) process motor and cognitive information. Their activity is affected by Parkinson's disease, in which dopamine concentration is decreased and acetylcholine concentration is increased. Acetylcholine activates muscarinic receptors in SPNs. Its main source is the cholinergic interneuron that responds with a briefer latency than SPNs during a cortical command. Therefore, an important question is whether muscarinic G-protein coupled receptors and their signaling cascades are fast enough to intervene during synaptic responses to regulate synaptic integration and firing. One of the most known voltage dependent channels regulated by muscarinic receptors is the KV7/KCNQ channel. It is not known whether these channels regulate the integration of suprathreshold corticostriatal responses. Here, we study the impact of cholinergic muscarinic modulation on the synaptic response of SPNs by regulating KV7 channels. We found that KV7 channels regulate corticostriatal synaptic integration and that this modulation occurs in the dendritic/spines compartment. In contrast, it is negligible in the somatic compartment. This modulation occurs on sub- and suprathreshold responses and lasts during the whole duration of the responses, hundreds of milliseconds, greatly altering SPNs firing properties. This modulation affected the behavior of the striatal microcircuit.

The channels that greatly modify the firing properties of SPNs are modulated by muscarinic receptors, and one of those channels are the M-channels (encoded by K V 7.2-K V 7.5/KCNQ2-KCNQ5 genes) controlled by membrane voltage. In spite of their relatively small contribution at the soma compartment, these channels modulate membrane potential from subthreshold to suprathreshold ranges and control input resistance, action potential threshold, and excitability [32][33][34][35]. Moreover, novel types of plasticity have been disclosed [36]. One hypothesis is that many of these actions can be manifested during corticostriatal responses and regulate firing properties during synaptic integration, but this has not been shown. Therefore, here, we investigated the role of 2 Neural Plasticity K V 7/M channels in the corticostriatal synaptic integration of GABAergic SPNs in vitro by using selective agonist and antagonist (retigabine and XE991, resp.). We found that current carried by K V 7 channels and regulated by muscarinic receptors greatly modifies the firing properties of GABAergic projection neurons during suprathreshold responses. Moreover, this firing modulation affects the behavior of the whole striatal microcircuit [37].

Slice Preparation.
The protocols followed the National University of Mexico guide for the care and use of laboratory animals (CICUAL-EGP41- 14) including minimizing the number of animals to achieve statistical significance and the avoidance of animal suffering. D 1 and D 2 dopamine receptor eGFP BAC transgenic mice were used, between postnatal days 20-35 (developed by the GENSAT). Wild mice and nonfluorescent cells of BAC-mice were also recorded to detect possible inconsistencies due to transgenes expression. The animals were anesthetized with ketamine/xylazine. Their brains were quickly removed and placed into ice cold bath saline containing (in mM): 126 NaCl, 3 KCl, 25 NaHCO 3 , 1 MgCl 2 , 2 CaCl 2 , 11 glucose, 300 mOsm/L, pH 7.4 with 95% O 2 , and 5% CO 2 . Hemispheres were separated and parasagittal corticostriatal slices (250-300 m thick) were cut using a vibratome and stored in oxygenated bath saline at room temperature. Recordings were carried out in the dorsal striatum. Stimulation was performed with concentric bipolar electrodes (tip = 50 m) located in the cortex, as previously described [1]. After recordings, neurons were injected with biocytin and merged with eGFP-positive visualization or else, immunoreacted for ChaT to observe on a confocal microscope as previously described [8].

Current Clamp Recordings.
We recorded from sagittal brain slices of BAC D1 or 2 eGFP transgenic mice. Slices were submerged in an iced saline solution containing (in mM): 124 NaCl, 2.5 KCl, 1.3 MgCl 2 , 2 CaCl 2 , 26 NaHCO 3 , 1.2 NaH 2 PO 4 , and 15 glucose (pH = 7.4, 300 mOsm/L, saturated with 95% O 2 and 5% CO 2 ). They were left for equilibration in this saline at room temperature for about 1 h. Single slices were transferred to a submerged recording chamber and superfused continuously with oxygenated saline (2-3 mL/min). Current-clamp recordings were performed with the patch clamp technique in the whole cell configuration in SPNs from the dorsal striatum. The slices were visualized using infrared differential interference contrast (IR-DIC) microscopy with an upright microscope and a digital camera. Data acquisition used software designed in the LabVIEW environment (National Instruments, Austin TX). Patch pipettes (3-6 MΩ) were filled with internal saline containing (in mM): 115 KH 2 PO 4 , 2 MgCl 2 , 10 HEPES, 1.1 EGTA, 0.2 ATP, 0.2 GTP, and 5% biocytin (pH = 7.2; 285 mOsm/L). In some experiments perforated patch clamp microelectrodes were used. No substantial differences between these configurations and previous recordings obtained with intracellular recordings were noted. Internal solution containing (in mM): 150 KCl, 10 HEPES, and final pH 7.2 and 280 mOsm/L was used. Stock solution of amphotericin B (66 g/mL) in dimethyl sulfoxide was diluted in the perforated patch internal solution for a final concentration of 180 g/mL. Transmembrane current was monitored continuously by applying a 10-20 mV pulse, from a holding potential of −80 mV.
Corticostriatal suprathreshold responses were evoked and recorded by stimulating sensory-motor cortical areas with concentric bipolar electrodes (50 m at the tip; FHC, Bowdoinham, ME). The distance between recording and stimulating electrodes was about 1 mm. Synaptic responses were evoked by a series of current pulses of increasing intensities until eliciting suprathreshold responses, with or without the firing of repetitive action potentials [8]. The stimuli were produced by the program but controlled by an isolation unit (Digitimer Ltd., England). The membrane potential was held at about −80 mV (near the "down"-state or resting membrane potential; −81 ± 5 mV; = 24) while polysynaptic corticostriatal responses, lasting hundreds of milliseconds [1,8], were induced.

Voltage-Clamp
Recordings. Synaptic currents were evoked with field stimulation. The field electrode was positioned in the cortex: a bipolar concentric tungsten electrode (50 m at the tip). Paired stimuli were used to test the interference of presynaptic muscarinic receptors (20 ms of interstimulus interval; 0.2-0.4 ms duration; 1-40 V delivered through the stimulating electrode; at a frequency of 0.1 Hz). These experiments were carried out in presence of bicuculline (10 M). Traces shown are the average of 2 min recordings (10 traces) taken once the amplitude had been stabilized in a given condition. A small hyperpolarizing voltage command (10 mV) was constantly given during the experiment to monitor input conductance.

Calcium
Imaging. These methods have been described before [38]. Briefly, mice were transcardially perfused with an ice-cold solution containing (in mM): 234 sucrose, 2.5 KCl, 7 MgCl 2 , 0.4 CaCl 2 , 28 NaHCO 3 , 1.44 NaH 2 PO 4 , 7 glucose, 0.28 ascorbic acid, and 4.5 pyruvate (pH = 7.4 with NaOH, saturated with 95% O 2 -5% CO 2 ) before decapitation. Slices were then obtained with the same procedure as above. They were incubated in the dark for 40 min with 6.5 M fluo-4 AM (Invitrogen, Life Technologies) and equilibrated with 95% O 2 -5% CO 2 . Slices were then superfused with control saline in a chamber located on the stage of an upright microscope equipped with a 20x water-immersion objective (Olympus XLUMPlanFI; Olympus America Inc.). Excitation at 565 nm was performed with a Lambda LS illuminator (Sutter Instruments, Novato, CA). Experiments were performed at room temperature. Images were acquired with a cooled digital camera (CoolSNAP K4, Photometrics; Roper Scientific, Tucson, AZ) at 100-250 ms/frame. Data acquisition software was also designed in the LabVIEW environment. The imaged field was 800 × 800 m. Short movies (180 s and 20 ms exposure) were taken at different pharmacological conditions. The same program performed preliminary image processing. All active neurons in a field were automatically identified and their mean fluorescence was measured as a function of time. Single pixel noise was discarded using a 5-pixel ratio mean filter. Calcium dependent fluorescence signals were computed as ( − / ), where is fluorescence intensity at any frame and is resting fluorescence. Calcium signals elicited by action potentials were detected based on a threshold value given by the first time derivative of their calcium transients (2.5 SD of the noise) [38]. Calcium transients were signaled by dots in a raster plot where each row represented the activity of one neuron and the -axis represents time. Summed activity was graphed below the raster plot in a form of histogram. 10,000 Monte Carlo simulations were used to find the significance of neurons being active together. In this way we could follow the activity of dozens of neurons with single cell resolution [38].

Muscarinic Actions on Corticostriatal Responses.
The striatum is mainly composed of GABAergic neurons: about 90-95% are striatal projection neurons (SPNs) and about 5-10% are interneurons, most of them being GABAergic. In addition, this nucleus is extremely rich in acetylcholine (ACh), cholinergic receptors, and cholinergic interneurons [16,17,[39][40][41]. Several classes of striatal interneurons activate slightly before or in correlation with SPNs following a cortical stimulus. A suprathreshold stimulus may activate SPNs directly and indirectly through the polysynaptic activation of interneurons and other SPNs [1,42]. Besides activating glutamatergic and GABAergic receptors, polysynaptic responses last hundreds of milliseconds and include the activation of muscarinic receptors as well as several classes of intrinsic voltage dependent currents [24,26,43,44]. This multisynaptic and convergent activation is one origin of "down"-to "up"-states voltage transitions [7]. Figure 1 shows typical firing modes of three double labeled and identified striatal neurons upon cortical stimulation: cholinergic (Figure 1(a)), a D 1 -receptor expressing direct pathway striatal projection neuron (dSPN) (Figure 1(b)), and a D 2 -receptor expressing indirect pathway striatal projection neuron (iSPN). It has been shown that cholinergic neurons respond with a slightly briefer latency than the responses of SPNs [1,24]. In addition, continuous firing of cholinergic interneurons maintains a tonic level of ACh in the striatum [7,24,45], muscarinic M 1 receptors are expressed in all SPNs, and K V 7 channels have been shown to compose a minor but functionally important part of the intrinsic voltage gated currents that are present in all SPNs [34].
Here, we show evidence ( Figure 2) that the response of SPNs to the same cortical stimulus is affected by activating muscarinic receptors. The same results can be obtained with perforated or nonperforated whole-cell recordings as well as with intracellular recordings [8]. Thus, Figures 2(a) and 2(b) show that the depolarization evoked by cortical stimulation was reduced in both classes of SPNs by blocking the activation of M 1 receptors by the very selective mamba toxin 7 (50 nM MT-7) [46,47]; indicating that G-protein coupled signaling activated by endogenous ACh was necessary to attain the level of depolarization to achieve repetitive firing [18]. After blockade of muscarinic M 1 class receptors the area under the synaptic response of dSPNs decreased by 23% (from 16,510 ± 1,495 mVms to 12,690 ± 1,218 (mVms); * * * < 0.0005; = 12; Figure 2(e)) and the same actions were revealed for iSPNs: MT-7 decreased the area under the synaptic response by 22% (from 11,360 ± 809 mVms to 8,891 ± 853 mVms; * * < 0.0005; = 9; Figure 2(e)). In both cases, firing was severely affected.
The reverse experiments are shown in Figures 2(c) and 2(d): muscarinic M 1 class receptors were activated by using the agonist muscarine (1-10 M): in dSPNs the response implied a larger depolarization reflected by the area under the synaptic response that increased 14% (from 17,290 ± 1,155 mVms to 19,690 ± 1,811 mVms; * * < 0.008; = 8; Figure 2(f)). The synaptic response of iSPN also increased for the same stimulation intensity by 32% (from 11,360 ± 983 mVms to 15,010 ± 1,203 mVms; * * < 0.002; = 10; Figure 2(f)). Modulation lasted the whole duration of the responses and it does not appear to be saturated since muscarinic actions could add to the endogenous ACh actions. Apparently, the activation of muscarinic receptors may produce changes in passive properties (e.g., electrotonic decay and membrane resistance) in the membrane compartment where most synaptic inputs are generated (mostly secondary and tertiary dendrites) and on intrinsic currents that become activated during synaptic suprathreshold depolarization [24,26] to explain the changes in amplitude of these responses.
Thus, previously, we have reported other muscarinic actions on the corticostriatal responses, for instance, the boosting of synaptic responses by facilitating Ca 2+ -currents. But these actions were due to muscarinic M 4 class receptors that appeared to act only in dSPNs [24]. In addition, modulation of Ca 2+ -activated K + -currents shunts the trains of action potentials generated in iSPNs making them briefer than those generated in dSPNs [26]. Therefore, the responses now described are the first to affect both classes of SPNs in the same way and with the same relative magnitude, affecting their whole duration. The obvious candidates to explain these responses are the K V 7/KCNQ channels classical effectors of muscarinic receptors, because it has been demonstrated that activation of muscarinic receptors closes K V 7 channels in these cells [34]. Moreover, single-cell reverse transcriptase-PCR confirmed the expression of KCNQ 2,3,5 mRNAs in SPNs, although their contribution to whole cell K + -current is relatively small [34]. Therefore, we next evaluated the action of these channels on the suprathreshold response.  Neural Plasticity pharmacological tools (Figures 3 and 4). First, we tested the response of iSPNs after single pulses of increasing intensity to evoke subthreshold, threshold, and suprathreshold responses, before (green traces are control records) and during 10-20 M XE991, a K V 7 channel antagonist. Similarly to the case of muscarinic M 1 class receptor blockade, the closing of K V 7 channels enhanced the evoked depolarization at all tested strengths of stimulation (black traces during XE991; Figure 3(a)). This is expected since XE991 is a K V 7 channel blocker that decreases membrane conductance thus boosting synaptic responses. In the inset of Figure 3(a) it is shown that autoregenerative calcium potentials ( = 9 out of 11 neurons) could be evoked after XE991 facilitation of the synaptic response [8,48,49], suggesting that these channels are necessary to control this outcome. Figure 3 Figure 4(e): one way ANOVA with Dunnett's multiple comparison test indicated that differences between XE991 and retigabine with the controls were significant ( * < 0.05).

K 7 Actions on Corticostriatal Responses in Both Classes
During single recordings in cell-focused studies there are often sources of variation such as the position of the electrodes in each experiment, whether the recorded cell was a main target of the cortical afferents being activated, and the activity of the microcircuit itself when each neuron has a role in a reverberant type of activity [21]. Therefore, to reinforce the statistical value of this findings we performed experiments where the activity of several of neurons were recorded by means of Ca 2+ imaging using fluo-4 [38]. Figure 5(a) illustrates one example of how cortically evoked intracellular Ca 2+ transients augmented by the presence of XE991 in one cell [28]. Figure 5(b) shows a raster type of plot where dots represent intracellular Ca 2+ transients as those in Figure 5(a) (cell activity), the -axis denotes time, and each row of the -axis represents activity of a single neuron. There was more evoked activity during XE991 than in the control ( = 6 slices from different animals). The histogram of Figure 5(c) illustrates the summed activity of neurons in the raster plot where each gray column denotes the times of stimulation. Cortical stimulus evoked more peaks of significant coactive neurons in the presence of XE991, suggesting that the action of the cortex was facilitated by activating more neurons when K V 7 channels were blocked. Tukey plots in Figure 5(d) summarizes the statistics of this sample ( * * < 0.02; Mann Whitney's test).
The action mediated by K V 7 channels in the suprathreshold responses (according to pharmacological tools) is the first one that affected both classes of neurons during the whole duration of their responses. These responses are the most similar to those shown for M 1 receptors ( Figure 2) in sharp contrast with previous reports of muscarinic involvement in the synaptic corticostriatal response of SPNs, while M 4receptor action is only present in dSPNs [24] and Ca 2+activated K + -currents act differentially in dSPNs and iSPNs [26]. It is known that M 1 muscarinic receptors close K V 7 channels through the phosphorylation of PIP 2 [34]. And although differences reported for each type of SPNs remained (enhanced regenerative events in iSPNs; Figure 3(a); and more prolonged trains of spikes in dSPNs), the closing of K V 7 channels, acting on both cell classes, augments the circuit activity after a cortical command.

K 7 Blockade Evokes Down-and Up-State Voltage
Transitions. iSPNs have been posited as more excitable than dSPNs [50]. Surprisingly, however, K V 7 blockade with XE991 evoked "down"-and "up"-state voltage transitions more readily in dSPNs than in iSPNs as recorded in whole cell. Figure 6(a) shows that XE991 (10-20 M) enhanced the depolarization and duration of action potentials trains in dSPNs (red control; black with XE991). Commonly, these cells are silent without stimulation ( Figure 5(b)). However, in the absence of any overt stimulus, addition of XE991 into the bath saline evoked transitions between "down"-and "up"-states (Figure 6(c)), many of them sustaining trains of action potentials. This behavior was observed in 23% of recorded cells (3 out of 13) and could be observed for up to 45 min. This oscillatory behavior was similar to that evoked with NMDA in vitro [51].
By recording in small samples of cells, one cannot be sure that this action of XE991 is significant for the microcircuit. Therefore, in Figure 7 we show a representative experiment with Ca 2+ imaging and simultaneous recordings of a population of SPNs. Clearly, intracellular Ca 2+ transients increased in several cells (Figure 7(a)), and the raster plot (as that in Figure 5 but without cortical stimulation; Figure 7(b)) shows more neurons active during XE991 (neurons were sorted in ascending order to separate spontaneously active neurons from those recruited after XE991). Histogram of summed activity shows significant peaks of coactive cells only after XE991 (Figure 7  Total numbers of active cells were also significantly different (Figure 7(e); * * * < 0.003; Mann-Whitney's test). To conclude, these experiments demonstrate that the closing of a single class of K + channels, K V 7, not only affects scattered cells but the behavior of the circuit as a whole, even if they comprise a small fraction of whole cell K + current [34]. Analysis of the resultant circuit is out of the scope of the present work. But evidences that K V 7 channels are involved in the control of correlated firing exist [35,52]. These results suggest that the great increase in circuit activity during Parkinson's disease may be in part due to hypercholinergia [21]. in the presence of 10 M bicuculline and these were compared to responses obtained in the presence of XE991. Figures 8(a)  and 8(b) show a control recording in a dSPN (average in color and quantal variation in thin grey lines) and a recording in the presence of XE991, respectively. The superimposition is in Figure 8(c). There was a small decrease in current amplitude during the experiment as observed from the soma, suggesting again a decrease in membrane conductance in the region where the synaptic responses are generated (Figure 8(d)); however, there was no change in the paired pulse ratio (PPR; Figure 8(e)). Lack of significance in PPR changes is summarized in Figure 8(f) ( = 8). A similar experiment was performed in a sample of iSPNs (Figures 8(g)-8(l)). Here, the decrease in EPSC amplitude was larger suggesting that propagation in iSPNs dendrites is more important than in dSPNs dendrites, given that dendrites form the cell compartment where most synaptic inputs are generated [48,50].

Actions of K 7 Channels
Nonetheless, changes in PPR were not significant ( Figures  8(k) and 8(l)). It was concluded that XE991 reduced EPSCs amplitude in both SPNs without changing PPR and therefore, most actions observed were postsynaptic. In addition, it is known that M 2−4 receptors, not M 1 receptors, are located on presynaptic cortical glutamatergic terminals [10][11][12][13]53].

Actions of K 7 Channels Minimally Affect the Somatic
Compartment. As said before, it is known that most excitatory synapses in SPNs target the dendritic/spines compartment [48,50,[54][55][56]. Therefore, the suprathreshold corticostriatal responses described here were most probably generated in the dendritic compartment and recorded in the somatic compartment. But are K V 7 channels distributed equally in all neuronal membrane? To answer this question we asked what would be the influence of K V 7 channels in responses evoked at the somatic compartment. Figures 9(a)-9(d) show that there were changes in excitability produced in SPNs by XE991 after somatic current injection, although it was more effective at iSPNs. However, actions of Ca 2+ -activated K + -currents can also be seen on suprathreshold synaptic responses [26]. Then, point somatic voltage-clamp current-voltage relationships ( -plots) were explored in the voltage-clamp mode (Figures 9(e)-9(h); unclamped action currents were clipped). Black dots in Thus, although it is well established that dendrites of SPNs express multiple types of potassium channels that contribute to the complexity of neuronal discharge, the functional role of K V 7 channels during the suprathreshold corticostriatal response of SPNs had not been demonstrated before. Here this functional role is strongly suggested due to the use of selective pharmacological tools. Multiple neurotransmitters have been shown to down-or up-modulate K V 7 channels [57,58]. Therefore, this may be a way in which cholinergic innervation controls SPNs firing and circuits. In contrast to these results, K V 7 channels selectively influence somatic but not dendritic synaptic integration in pyramidal cells from the hippocampus [49,[59][60][61], although they control synaptic integration in pyramidal cortical neurons [62]. However, in similarity with pyramidal cells, SK-channels also play an important role in SPNs [26] and their interaction needs further study, for example, [63].
Up to now we have shown that K V 7 channels may influence synaptic integration because their selective blocker and opener acted on suprathreshold synaptic responses. Also, experiments in Figure 2 showed that a M 1 muscarinic receptor antagonist and agonist had very similar actions, respectively; they acted similarly in both classes of projection neurons, influenced firing, and their actions lasted during the whole response. Actions such as these had not been observed for other muscarinic effects [24,26]. To further show these similarities we performed occlusion experiments ( Figure 10). However, as mentioned before, muscarinic receptors have many actions on SPNs, on all classes of membrane currents: Na + , Ca 2+ , and K + [12, 17-21, 23-26, 30, 34, 44]. Therefore, occlusion experiments cannot be complete. However, if occlusion of these actions is large enough and differences can have a reasonably explanation that could be confirmed experimentally in the future, it could become very suggestive. Figure 10(a) shows that muscarine (1 M) had an additional action after XE991 (20 M), but this does not happen in iSPNs where occlusion is complete (Figure 10(b)). This can be easily explained by the actions of M 4 -receptors in dSPNs which are not present in iSPNs [24]. On the other hand, addition of XE991 in a response of a dSPN that provoked a few action potentials after muscarine obtained only a small additional depolarization. This increased number of action potentials fired shows that actions do not need to be large to influence firing (Figure 10(c)). It is rare that a receptor action saturates any effector, but comparing panels (a) and (c) reinforces the argument of this work and of a previous one [24]. A similar phenomenon occurred in an iSPN whose depolarization did not evoke an autorregenerative response after muscarine but attained it after addition of XE991 (Figure 10(d)). While these experiments ( = 6 for each sample) confirmed that muscarinic receptors have various actions on SPNs, the amount of depolarization added to the responses after sequential activation with muscarine and XE991 in either order suggests that a great part of the action may be due to K V 7 channels. Since K V 7 channels are modulated by a variety of neurotransmitters and intracellular signaling molecules [64][65][66], they provide an exquisite mechanism to fine-tune synaptic convergent integration from the sub-to suprathreshold ranges. It is thought that convergent and prolonged suprathreshold inputs provoke the "down"-to "up"-states voltage transitions [7,50,51,64] characteristic of SPNs firing. It is also known that during these transitions SPNs  [24]. (b) This additional action does not happen in iSPNs since they lack M 4 -receptors. (c) After muscarine, XE991 only provoked a small additional depolarization, since it is rare that a receptor action saturates completely a given effector. However, comparing (a) and (c) shows that additional action in (a) suggests that it is not due to M 1 -receptors [24]. (d) Additional action of XE991 is due to an autorregenerative response, which can also appear without it (cf. (b)).
become involved in correlated firing and network activity such as cell-assembly reverberations [38]. Finally, previous work had shown that a cortical stimulus may provoke prolonged synaptic responses in cholinergic interneurons and SPNs quasisimultaneously [1]. But to respond to released ACh, the receptors expressed by SPNs are G-protein coupled (muscarinic). In particular, modulation of K V 7 channels involves phosphatidylinositol 4, 5 biphosphate depletion [58,[67][68][69]. Thus, whether the signaling cascades involved are fast enough to modulate these complex synaptic responses during their whole duration was an open question. In this work we demonstrate that muscarinic signaling is definitively involved in the synaptic integration of SPNs and that this modulation affects the firing of these GABAergic neurons.

Conclusions
By activating SPNs and cholinergic interneurons almost simultaneously, acetylcholine modulates, through muscarinic receptors, the suprathreshold synaptic integration in striatal projection neurons. In a similar way, K V 7 channels act as "gain control" regulators of the synaptic response during its whole duration, in both classes of SPNs. Thus, blocking of M 1 class of muscarinic receptors decreases the responses and greatly abolishes firing, disclosing the action of endogenous acetylcholine [24]: the regulated closing of these channels during the response. Addition of the agonist muscarine does the opposite; it facilitates synaptic depolarization and firing. This last action is potently reproduced by the K V 7 channel blocker XE991 in both classes of neurons. Conversely, the K V 7 channel opener, retigabine, mimicked the action of the M 1 class receptor antagonist; it reduced the responses in both classes of SPNs. It has been demonstrated that muscarinic agonists close K V 7 channels [34]. Calcium imaging experiments showed that the efficiency of a cortical stimulus to recruit sets of coactive SPNs is increased when K V 7 channels are closed. Therefore, modulation of these channels not only enhances the response of scattered SPNs but facilitates their working together. In fact, XE991 alone could produce the appearance of "down"-to "up"-state transitions in SPNs. This action was not a random occurrence since it generated a definite increase in microcircuit activity. By inference, we show that the actions of K V 7 channels are postsynaptic and that they may occur in the dendritic compartment where most synaptic inputs are generated, since their action in somatically evoked responses were minimal. Finally, occlusion was almost complete when XE991 and muscarine are given together in spite of the various muscarinic actions, except when muscarine was given after XE991, suggesting the action of M 4 receptors [24]. In summary, the present results suggest that antimuscarinic therapy in Parkinson's disease and L-DOPA induced dyskinesia should be more selective and focused on M 1 -class receptors and clinical assays of Parkinsonian-dyskinetic patients using retigabine are scarce. Further basic and clinical research is necessary to better support this suggestion [70].