Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 2016 Dec 17;595(5):1711–1723. doi: 10.1113/JP273627

A unifying hypothesis for M1 muscarinic receptor signalling in pyramidal neurons

Sameera Dasari 1, Corey Hill 1, Allan T Gulledge 1,✉
PMCID: PMC5330881  PMID: 27861914

Abstract

Key points

  • Phasic release of acetylcholine (ACh) in the neocortex facilitates attentional processes.

  • Acting at a single metabotropic receptor subtype, ACh exerts two opposing actions in cortical pyramidal neurons: transient inhibition and longer‐lasting excitation.

  • Cholinergic inhibitory responses depend on calcium release from intracellular calcium stores, and run down rapidly at resting membrane potentials when calcium stores become depleted.

  • We demonstrate that cholinergic excitation promotes calcium entry at subthreshold membrane potentials to rapidly refill calcium stores, thereby maintaining the fidelity of inhibitory cholinergic signalling.

  • We propose a ‘unifying hypothesis’ for M1 receptor signalling whereby inhibitory and excitatory responses to ACh in pyramidal neurons represent complementary mechanisms governing rapid calcium cycling between the endoplasmic reticulum, the cytosol and the extracellular space.

Abstract

Gq‐coupled M1‐type muscarinic acetylcholine (ACh) receptors (mAChRs) mediate two distinct electrophysiological responses in cortical pyramidal neurons: transient inhibition driven by calcium‐dependent small conductance potassium (‘SK’) channels, and longer‐lasting and voltage‐dependent excitation involving non‐specific cation channels. Here we examine the interaction of these two cholinergic responses with respect to their contributions to intracellular calcium dynamics, testing the ‘unifying hypothesis’ that rundown of inhibitory SK responses at resting membrane potentials (RMPs) reflects depletion of intracellular calcium stores, while mAChR‐driven excitation acts to refill those stores by promoting voltage‐dependent entry of extracellular calcium. We report that fidelity of cholinergic SK responses requires the continued presence of extracellular calcium. Inhibitory responses that diminished after repetitive ACh application at RMPs were immediately rescued by pairing mAChR stimulation with subthreshold depolarization (∼10 mV from RMPs) initiated with variable delay (up to 500 ms) after ACh application, but not by subthreshold depolarization preceding mAChR stimulation. Further, rescued SK responses were time‐locked to ACh application, rather than to the timing of subsequent depolarizing steps, suggesting that cholinergic signal transduction itself is not voltage‐sensitive, but that depolarization facilitates rapid cycling of extracellular calcium through the endoplasmic reticulum to activate SK channels. Consistent with this prediction, rescue of SK responses by subthreshold depolarization required the presence of extracellular calcium. Our results demonstrate that, in addition to gating calcium release from intracellular stores, mAChR activation facilitates voltage‐dependent refilling of calcium stores, thereby maintaining the ongoing fidelity of SK‐mediated inhibition in response to phasic release of ACh.

Keywords: acetylcholine, calcium, muscarinic receptor, neocortex, pyramidal Neuron, SK channel

Key points

  • Phasic release of acetylcholine (ACh) in the neocortex facilitates attentional processes.

  • Acting at a single metabotropic receptor subtype, ACh exerts two opposing actions in cortical pyramidal neurons: transient inhibition and longer‐lasting excitation.

  • Cholinergic inhibitory responses depend on calcium release from intracellular calcium stores, and run down rapidly at resting membrane potentials when calcium stores become depleted.

  • We demonstrate that cholinergic excitation promotes calcium entry at subthreshold membrane potentials to rapidly refill calcium stores, thereby maintaining the fidelity of inhibitory cholinergic signalling.

  • We propose a ‘unifying hypothesis’ for M1 receptor signalling whereby inhibitory and excitatory responses to ACh in pyramidal neurons represent complementary mechanisms governing rapid calcium cycling between the endoplasmic reticulum, the cytosol and the extracellular space.

Abbreviations

aCSF

artificial cerebral spinal fluid

EK

equilibrium potential for potassium

ER

endoplasmic reticulum

IP3

inositol trisphosphate

ISF

instantaneous spike frequency

mAChR

muscarinic acetylcholine receptor

NCX

sodium–calcium exchanger

OGB6

Oregon Green 488 BAPTA‐6F

PMCA

plasma membrane calcium ATPase

PLC

phospholipase C

RMP

resting membrane potential

SERCA

sarco/endoplasmic reticulum calcium ATPase

Introduction

Pyramidal neurons expressing M1‐type muscarinic acetylcholine (ACh) receptors (mAChRs) coupled to Gq‐type G‐protein alpha subunits are key targets of cholinergic afferents to the cerebral cortex and hippocampus (Gulledge et al. 2009; Dasari & Gulledge, 2011). M1 receptors initiate two distinct responses in cortical pyramidal neurons: transient inhibition induced by inositol trisphosphate (IP3)‐triggered calcium release from intracellular stores and subsequent activation of calcium‐dependent small conductance potassium (‘SK’) channels, and a longer lasting and voltage‐dependent excitation mediated primarily by one or more non‐specific cation channels (Andrade, 1991; Haj‐Dahmane & Andrade, 1996; Klink & Alonso, 1997; Haj‐Dahmane & Andrade, 1998; Egorov et al. 2002; Shalinsky et al. 2002; Egorov et al. 2003; Magistretti et al. 2004; Yan et al. 2009; Zhang & Seguela, 2010; Rahman & Berger, 2011; Zhang et al. 2011; Dasari et al. 2013; Lei et al. 2014).

SK‐mediated inhibitory responses to ACh ‘run down’ rapidly during repetitive M1 receptor stimulation at resting membrane potentials (RMPs), but remain robust when neurons are depolarized (Gulledge & Stuart, 2005). Because the ongoing fidelity of inhibitory cholinergic responses at depolarized potentials is sensitive to extracellular cadmium (Gulledge & Stuart, 2005), a non‐specific blocker of calcium conductances, we hypothesized that response rundown at RMPs reflects depletion of intracellular calcium stores that are refilled by voltage‐dependent influx of extracellular calcium. Further, because mAChR stimulation facilitates the recovery of inhibitory cholinergic responses (Gulledge et al. 2007), we further hypothesized that M1 receptors engage two complementary signal transduction pathways in pyramidal neurons, one acting to promote IP3‐dependent calcium release from the endoplasmic reticulum (ER) to activate SK channels, and another that enhances action potential output in response to suprathreshold excitatory input, while at the same time promoting voltage‐dependent refilling of ER calcium stores. According to this ‘unifying hypothesis’ of M1 cholinergic transmission in cortical pyramidal neurons, cholinergic excitation reflects, in part, inward calcium currents that refill the intracellular calcium stores responsible for SK‐mediated inhibition. Here we test these hypotheses, confirming that activation of mAChRs facilitates the entry of extracellular calcium at subthreshold membrane potentials, and that this calcium entry is necessary to maintain the ongoing fidelity of cholinergic inhibitory responses over time.

Methods

Ethical approval

Experiments were performed on 5‐ to 8‐week‐old C57BL/6 mice of both sexes according to procedures approved by the Institutional Animal Care and Use Committee of Dartmouth College. Mice, originally sourced from the Jackson Laboratory, were bred in‐house and maintained on a 12 h/12 h light–dark cycle with unlimited access to food and water in facilities accredited by the Association for Assessment and Accreditation of Laboratory Animal Care.

Slice preparation and recording

On the day of experiments, animals were anaesthetized with vaporized isoflurane (2%) and decapitated, with brains removed into artificial cerebral spinal fluid (aCSF) containing (in mm): 125 NaCl, 25 NaHCO3, 3 KCl, 1.25 NaH2PO4, 0.5 CaCl2, 5 MgCl2 and 25 glucose (bubbled with 95% O2/5% CO2). Coronal brain slices of the medial prefrontal cortex (250 μm thick) were cut using a vibratome and stored in aCSF containing 2 mm CaCl2 and 1 mm MgCl2 (continuously bubbled with 95% O2/5% CO2). For the majority of experiments, recording pipettes (5–8 MΩ) were filled with a pipette solution composed of (in mm): 140 potassium gluconate, 2 NaCl, 2 MgCl2, 10 Hepes, 3 Na2ATP and 0.3 NaGTP (pH 7.2 with KOH). For perforated‐patch recordings (see Fig. 2 B), recording pipettes were filled with (in mm): 125 KCl, 25 glucose, 15 Hepes, 10 potassium gluconate, 3 MgCl2 and 0.1 Lucifer yellow (pH 7.2 with KOH). Immediately before perforated‐patch experiments, gramicidin was dissolved in DMSO and diluted into the pipette saline to a final concentration of 20 μg ml–1. Perforated‐patch recordings utilized unfilamented patch pipettes that were front‐filled with gramicidin‐absent internal saline, and back‐filled with internal saline including gramicidin (Gulledge & Stuart, 2003). Recordings were made from visually identified layer 5 pyramidal neurons in the prelimbic region of the medial prefrontal cortex. Data were acquired with Axograph software (Axograph, Sydney, Australia) using a BVC‐700 amplifier (Dagan Corporation, Minneapolis, MN, USA) and an ITC‐18 digitizer (HEKA Instruments, Bellmore, NY, USA). Voltage responses were filtered at 5 kHz, digitized at 25 kHz, and corrected for the liquid junction potentials for potassium gluconate (+12 mV) and gramicidin‐based (+2 mV) pipettes. For traditional whole‐cell recordings, series resistances (10–30 MΩ) were maximally compensated. For perforated‐patch recordings, series resistances were generally greater than 100 MΩ, and were left uncompensated. The integrity of perforated‐patch recordings was monitored visually, and neurons exhibiting somatic leak of Lucifer yellow were discarded. All experiments were carried out at 35 ± 1°C.

Figure 2. State‐dependent rundown of cholinergic responses does not reflect reduced sensitivity of IP3 receptors or intracellular dialysis during whole‐cell recording.

Figure 2

Open in a new tab

A1, responses to three progressively longer applications of ACh (100, 500 and 1000 ms) occurring during action potential generation (red) or at the RMP (−80 mV, black). A2, summary comparison of the mean durations of spike inhibition (red) and amplitudes of SK responses (black) in eight neurons. B1, responses to ACh at the RMP (−84 mV; black) and during action potential generation (red) during gramicidin‐based perforated‐patch recording of a layer 5 pyramidal neuron. B2, summary comparison of the mean durations of spike inhibition (red) and amplitudes of SK responses (black) in seven neurons recorded with gramicidin.

To measure cytosolic calcium changes, 100 μm Oregon Green 488 BAPTA‐6F (OGB6; Life Technologies, Carlsbad, CA, USA) was included in the patch pipette, and images (∼20 Hz) were acquired with a CCD camera (Photometrics, Tucson, AZ, USA; 470 nm LED illumination) on a conventional upright epifluorescence microscope, or with a two‐photon laser scanning microscope (Brukur Ultima) using 820 nm excitation from a Ti:sapphire laser (Coherent, Santa Clara, CA, USA). Mean somatic florescence intensities were background subtracted and signals presented as the ratio of the change in mean fluorescence (relative to baseline mean fluorescence) divided by baseline mean fluorescence (∆F/F).

To stimulate cholinergic responses, ACh was dissolved in aCSF (to 100 μm) and focally applied from a patch pipette near the somata of recorded neurons (∼10 psi and for a duration of 100 ms, unless otherwise noted). When ACh was applied at RMPs, neurons were initially ‘primed’ with suprathreshold current injections (∼300 pA; 5.7 s) to maximize initial responses to ACh (see Gulledge et al. 2007, and Fig. 2 A). To apply ACh in the presence of nominally calcium‐free aCSF (3 mm Mg2+), ACh dissolved in calcium‐free aCSF was loaded into one channel of a two‐channel theta‐glass patch pipette, while the other channel contained calcium‐free aCSF alone. Each of the two pipette channels was independently controlled, allowing transient application of ACh alone, or application of ACh during longer applications of calcium‐free aCSF.

The magnitude of cholinergic inhibitory responses was measured as the amplitude of hyperpolarizing responses from RMPs, or as the duration of action potential inhibition during which ACh was delivered during periods of constant suprathreshold current injection. Excitatory responses to ACh during periods of action potential generation were quantified as the percentage change in mean instantaneous spike frequency (ISF) as measured during the 2 s before ACh application and over the first 2 s following the end of SK‐mediated inhibition (Gulledge et al. 2007).

Statistical analysis

Data are presented as mean ± SEM. Comparisons of paired data used a Student's t‐test (two‐tailed) for paired samples, or repeated measures ANOVA, as appropriate. Comparisons of multiple group means involved one‐way ANOVAs.

Results

To test the hypothesis that state‐dependent rundown of inhibitory cholinergic responses reflects a failure of mAChR‐induced calcium release from intracellular stores, we applied ACh (100 μm, 100 ms) three consecutive times (0.125 Hz) to layer 5 pyramidal neurons filled with the calcium indicator OGB6 (100 μm). During periods of current‐induced action potential generation (mean frequency = 14 ± 2 Hz; n = 8), focal ACh generated biphasic responses characterized by transient cessation of action potential generation (mean durations of inhibition were 1.0 ± 0.2, 0.9 ± 0.2 and 0.9 ± 0.2 s, respectively, for three consecutive ACh applications), followed by increases in action potential frequency above baseline levels of 167 ± 17, 153 ± 10 and 152 ± 10%, respectively. Coinciding with these electrophysiological responses were somatic calcium transients of 31 ± 4, 26 ± 3 and 22 ± 2% (∆F/F × 100; Fig. 1 A and B). On the other hand, when triggered at RMPs, inhibitory ‘SK’ responses (−2.6 ± 0.4 mV) and calcium transients (30 ± 4%) occurred only in response to the initial application of ACh. During the two subsequent presentations of ACh, inhibitory responses and calcium transients were absent (Fig. 1 A and B). These data suggest that single exposures to ACh are sufficient to deplete intracellular calcium stores, but that refilling of those stores is impaired at RMPs. However, since initial ACh‐driven calcium transients were of similar magnitude at both RMPs and during activity (30 ± 4 and 31 ± 4%, respectively; P = 0.88, paired Student's t‐test), it is likely that ACh generates a similar level of mAChR signal transduction in the two conditions. Finally, we tested for correlation between intracellular calcium signals and electrophysiological responses to initial applications of ACh delivered during periods of action potential generation or at RMPs (Fig. 1 C). While the enhancement of action potential frequency was not correlated with intracellular calcium transients (R 2 = 0.09, P = 0.47, linear regression), inhibitory SK responses occurring during active spiking (R 2 = 0.77, P < 0.05) or at RMPs (R 2 = 0.67, P < 0.05) were positively correlated with the magnitude of intracellular calcium transients.

Figure 1. State‐dependent rundown of ACh‐stimulated calcium release from intracellular stores.

Figure 1

Open in a new tab

A, voltage (top) and calcium (∆F/F; bottom) responses to three consecutive applications of ACh (100 μm; 100 ms) at RMPs (black; −70 mV) or during suprathreshold somatic current‐injection (red). Calcium transients and hyperpolarizing responses ‘rundown’ during repetitive applications of ACh at rest, but persist when ACh is delivered during current‐induced action potential generation. B, comparison of ACh‐stimulated calcium transients (as measured as changes in fluorescence) to three applications of ACh at RMPs (black) or during current‐induced activity (red). * P < 0.05. C, plots of percentage increases in spike rate (top) and durations of spike inhibition (middle) measured following ACh applications during current‐induced action potential generation, and SK‐mediated response amplitudes measured at RMPs (bottom), vs. coinciding changes in cytosolic calcium (∆F/F). Red lines indicate linear regressions to data.

The results above suggest that rundown of SK responses may reflect depletion of intracellular calcium stores following mAChR activation at RMPs. However, IP3 receptors are directly regulated by intracellular calcium concentrations, with optimal IP3 receptor sensitivity to IP3 occurring when calcium levels are at intermediate concentrations (∼200 nm), and with IP3 sensitivity reduced when calcium concentrations are lower or higher than this optimal level (Bezprozvanny et al. 1991). To test whether SK response rundown at RMPs reflects reduced IP3 receptor sensitivity following initial applications of ACh, we increased the duration of ACh application during the subsequent two applications (to 500 and 1000 ms, respectively) to enhance mAChR activation on each subsequent application (Fig. 2 A). Under these conditions, cholinergic SK responses continued to rundown at RMPs, with response amplitudes decreasing from −2.8 ± 0.3 mV following the first 100 ms ACh application, to −1.5 ± 0.4 and −0.3 ± 0.2 mV following the 500 and 1000 ms ACh applications, respectively (n = 8; P < 0.05, repeated measures ANOVA). At the same time, cholinergic inhibition of action potential genesis at depolarized potentials remained robust across all three ACh applications, with the mean durations of inhibition being 0.9 ± 0.1, 0.8 ± 0.1 and 0.7 ± 0.1 s, respectively (Fig. 2 A; P = 0.43, repeated measures ANOVA). These results suggest that SK response rundown at RMPs does not result from reduced IP3 receptor sensitivity.

Another possibility is that dialysis of the cytosol during whole‐cell recording may disrupt processes necessary for store‐refilling at RMPs. To control for this, we recorded seven neurons using a perforated‐patch configuration that preserves intracellular integrity (Fig. 2 B; see Methods). Under these conditions, SK responses at RMPs continued to rundown rapidly, from −3.6 ± 0.7 mV during the first ACh application to −0.5 ± 0.6 mV on each of the next two ACh applications (n = 7; P < 0.05, repeated measures ANOVA). In these same neurons, ACh continued to generate robust SK responses when neurons were subjected to suprathreshold depolarization (Fig. 2 B), with mean durations of inhibition following each of the three ACh applications being 0.7 ± 0.3, 0.6 ± 0.2 and 0.6 ± 0.2 s, respectively (P = 0.22, repeated measures ANOVA). Together, these data confirm that inhibitory cholinergic responses preferentially rundown at RMPs, while remaining robust when neurons are depolarized.

If rundown of SK responses at RMPs reflects depletion of intracellular calcium stores, how are stores refilled during periods of action potential generation? Because cholinergic inhibitory responses and their associated calcium transients are sensitive to extracellular cadmium (Gulledge & Stuart, 2005), we reasoned that store refilling must require calcium entry from the extracellular space. We tested this by challenging cholinergic inhibitory responses with local perfusion of nominally calcium‐free aCSF (replaced with equimolar Mg2+). Calcium‐free aCSF impaired both ACh‐induced calcium transients and inhibitory SK responses (Fig. 3). Under baseline conditions, when ACh was delivered at RMPs following priming depolarizing steps (Fig. 3 A), inhibitory SK responses were −3.7 ± 0.3 mV (n = 14). When calcium‐free aCSF was focally applied during and following priming steps, these responses were reduced to −1.2 ± 0.2 mV (P < 0.05, paired Student's t‐test; Fig. 3 A and B). In a second set of experiments, in which neurons were filled with the calcium indicator OGB6 (100 μm; n = 6; Fig. 3 C), local perfusion of calcium‐free aCSF reversibly reduced both cholinergic SK responses (from −2.8 ± 0.4 to −0.7 ± 0.2 mV; Fig. 3 D; P < 0.05, repeated measures ANOVA) and calcium transients (from 50 ± 9.0 to 6 ± 0.1%; Fig. 3 E; P < 0.05, repeated measures ANOVA).

Figure 3. Extracellular calcium is required for ACh‐triggered calcium transients and inhibitory voltage responses.

Figure 3

Open in a new tab

A, ACh delivered after a 5.7 s loading step in baseline conditions (black traces) generates an inhibitory response from the RMP (−74 mV). Cholinergic inhibition is abolished during focal application of calcium‐free aCSF (3 mm Mg2+; red traces). Inset: expanded responses to ACh. B, comparison of mean SK response amplitudes in baseline conditions (black) and during focal application of calcium‐free aCSF (red) in 14 neurons. C, in a separate group of neurons (n = 6), cholinergic responses from RMPs and somatic calcium transients were monitored in baseline conditions (black traces; RMP is −74 mV), during focal application of calcium‐free aCSF (green traces; RMP is −74 mV), or after recovery (blue traces; RMP is −75 mV). For each condition, top traces show voltage responses from RMPs, while lower traces show coinciding calcium transients. Focal application of calcium‐free aCSF occurred during the loading step (not shown) and during ACh application. D and E, comparisons of the peak amplitudes of cholinergic inhibitory voltage responses (D) and calcium transients (E) before (black), during (green) and after (blue) focal application of calcium‐free aCSF. * P < 0.05 (repeated measures ANOVA).

Local perfusion of calcium‐free aCSF also impaired inhibitory SK responses and calcium transients following ACh delivery during periods of sustained action potential generation (Fig. 4). Under baseline conditions, repetitive applications of ACh (0.125 Hz; dissolved in calcium‐free aCSF) reliably generated inhibitory responses and coincident intracellular calcium transients (Fig. 4 A). However, when calcium‐free aCSF was locally perfused between the first and final ACh applications, inhibitory responses and calcium transients were effectively eliminated (n = 7; Fig. 4 A and C). To control for potential interference of ACh signalling by focal application of calcium‐free saline, in additional experiments we paired ACh applications with local perfusion of normal (i.e. 2 mm Ca2+, 1 mm Mg2+) aCSF (n = 5), and observed no significant impact on cholinergic responses (Fig. 4 B and C). Together, these results demonstrate that fidelity of cholinergic inhibition over time requires entry of calcium from the extracellular space, and that this entry is promoted by depolarization from RMPs.

Figure 4. Extracellular calcium is necessary for repetitive generation of cholinergic SK responses during periods of action potential generation.

Figure 4

Open in a new tab

A, voltage responses (top) and calcium transients (bottom) to repetitive ACh application (100 μm, 100 ms, at 8 s intervals) during current‐induced action potential generation in baseline trials (blue traces), and in trials in which calcium‐free aCSF was applied (green traces). B, control experiments comparing voltage responses (top) and plots of instantaneous spike frequency (bottom) for ACh applications in baseline conditions (yellow) and during focal application of regular aCSF (red). C, comparisons of inhibitory ACh response durations (left) and amplitudes of ACh‐induced calcium transients (right) across the indicated experimental conditions. * P < 0.05 (one‐way ANOVA).

We previously reported that maintenance of repetitive cholinergic inhibitory responses is voltage‐dependent, but does not require action potential generation (Gulledge & Stuart, 2005). Further, we found that prior exposure to ACh enhanced voltage‐dependent recovery from response rundown (Gulledge et al. 2007). These observations suggest that pairing mAChR activation with subthreshold depolarization may be sufficient to rescue cholinergic inhibition. To test this hypothesis, we ran down inhibitory cholinergic responses with three sequential applications of ACh (0.33 Hz) at RMPs, and then applied a subsequent ‘test’ application of ACh 18 s later under a variety of experimental conditions (Fig. 5). When ACh was delivered at RMPs, cholinergic SK responses (initially −2.9 ± 0.2 mV during initial ACh applications; n = 17) ran down rapidly, being absent or very small (−0.6 ± 0.2 mV; n = 17) following the test application of ACh 18 s later (Fig. 5 A). Pairing the initial application of ACh with a subthreshold current step (2.5 s, generating 8.2 ± 0.7 mV of depolarization) enhanced initial SK response amplitudes to −5.9 ± 0.3 mV (P < 0.05, paired Student's t‐test), but did not prevent rundown of responses over time (mean SK response at RMPs 18 s later was −0.7 ± 0.2 mV; Fig. 5 B). While subthreshold depolarizations occurring alone, prior to the test application of ACh, failed to rescue cholinergic inhibitory responses (Fig. 5 C), test responses (−6.1 ± 0.6 mV) were rescued to baseline amplitudes (−6.3 ± 0.3 mV; P = 0.71, paired Student's t test) when test applications of ACh occurred concurrently with subthreshold depolarization (Fig. 5 D). Rescued SK responses during subthreshold depolarization could not be attributed to the increase (36 ± 4%; n = 17) in driving force for potassium conductances (E K = −102 mV), as SK responses to test ACh applications at depolarized potentials remained significantly larger (at −6.3 ± 0.3 mV) than predicted by the increased driving force alone (prediction = −1.1 ± 0.3 mV; P < 0.05, paired Student's t‐test).

Figure 5. SK responses are facilitated by coincident pairing of ACh with subthreshold depolarization.

Figure 5

Open in a new tab

A, three consecutive applications of ACh (3 s intervals; black and grey arrows) to a layer 5 pyramidal neuron at rest (RMP of −77 mV) result in ‘rundown’ of ACh‐induced SK responses that persists during a ‘test pulse’ of ACh applied 18 s later (red arrow). B, similar to A, but with the first application of ACh paired with a subthreshold current step. C, similar to B, but with a second subthreshold current injection occurring just prior to the test pulse (red arrow). D, similar to B, but with the ACh test application occurring concurrently with the subthreshold current step. The RMP for all trials (A–D) was −77 mV. E, comparison of the amplitudes of ACh‐induced SK responses following initial (black) and test (red) applications of ACh in the indicated experimental conditions (A–D). In all experiments, 4 s of suprathreshold current was applied shortly before beginning of the trial to prime initial SK responses. * P < 0.05.

How does subthreshold depolarization rescue cholinergic SK responses? One possibility is that M1 receptors are voltage‐sensitive (Ben‐Chaim et al. 2003, 2006; Rinne et al. 2015) and unable to reliably initiate signal transduction at RMPs. Alternatively, response rundown could reflect depletion of intracellular calcium stores that are rapidly refilled by coincident mAChR stimulation and subthreshold depolarization. To test these two possibilities, we compared the latencies of rescued cholinergic inhibitory responses when subthreshold depolarizing current steps were applied at variable delays (0–500 ms) relative to ACh application (Fig. 6). If rescue of cholinergic inhibition results from voltage‐sensitive M1 receptors, mAChR signal transduction, and therefore the latencies of rescued responses, should be time‐locked to subthreshold depolarization rather than to ACh exposure. Alternatively, if response rundown reflects insufficient calcium stores to engage SK channels, and response rescue depends upon voltage‐ and mAChR‐dependent store refilling, the latencies of rescued SK responses should be less dependent on current step timing.

Figure 6. Cholinergic SK responses are time‐locked to ACh exposure, rather than to the timing of subthreshold current steps.

Figure 6

Open in a new tab

A, superimposed voltage responses for trials in which ACh was applied three times (black and grey arrows) to ‘run down’ cholinergic SK responses, followed by a test application of ACh (red arrow) paired with a subthreshold current step applied at variable delay (0–500 ms, in 100 ms intervals). RMP was −75 mV on the first trial (black trace), and −74 mV on subsequent delayed test trials (grey traces). B, enlarged view of cholinergic responses to the test applications of ACh shown in A. C, plot of latencies to peak SK response for initial (‘baseline’) ACh applications (black) and for SK responses occurring after test ACh applications were paired with variably timed subthreshold current steps (red). The dashed line (45 deg) indicates the predicted delay in peak SK response if cholinergic signal transduction were time‐locked to current steps, rather than to ACh application. Black asterisks indicate P < 0.05 when comparing baseline response latencies to test response latencies. Red asterisks indicate P < 0.05 when comparing test response latencies to those predicted for voltage‐dependent mAChR signal transduction. D, plot of the amplitudes of cholinergic inhibitory responses during baseline (black) and test (red) ACh applications. Black asterisks indicate P < 0.05 when comparing test response amplitudes to baseline response amplitudes.

Latencies to peak inhibitory responses following baseline applications of ACh (i.e. those paired with subthreshold depolarization following suprathreshold priming steps) varied from 658 ± 58 ms during initial trials to 743 ± 55 ms on final trials (n = 17; P = 0.43, paired Student's t‐test). In these same trials, latencies of rescued inhibitory responses following test ACh applications 18 s later varied from 721 ± 60 ms in initial trials (where current steps were coincident with ACh applications) to 853 ± 36 ms when current steps were delayed by 500 ms (P < 0.05, Student's t‐test; Fig. 6 A–C). Importantly, for all current step delays beyond 100 ms, this increase in response latency was significantly smaller than predicted for voltage‐dependent mAChR signal transduction (Fig. 6 C; P < 0.05; paired Student's t‐test). In these same trials, response amplitudes following test applications of ACh were largely comparable to their baseline controls, except for a small decrease in amplitudes occurring when current steps were delayed by 400 or 500 ms (Fig. 6 D; P < 0.05, paired Student's t‐test). These results support the hypothesis that rundown of inhibitory SK responses reflects depletion of calcium stores, rather than a failure of mAChR signal transduction at RMPs.

If subthreshold depolarization rescues cholinergic inhibition by refilling intracellular calcium stores, response rescue should require the presence of extracellular calcium. To test this, we attempted to rescue previously rundown inhibitory responses by pairing ACh applications with subthreshold depolarization in baseline conditions (i.e. in the absence of local perfusion, n = 5, or with local perfusion of regular aCSF, n = 6) and after local perfusion with calcium‐free aCSF (n = 11; Fig. 7). Under baseline conditions, inhibitory cholinergic response amplitudes were −5.4 ± 0.6 mV, while rescued responses were −6.5 ± 0.7 mV (n = 11; P = 0.10, paired Student's t‐test). On the other hand, in trials where calcium‐free aCSF was locally perfused between baseline and test ACh applications, response amplitudes dropped from −5.8 ± 0.7 mV (baseline) to −2.2 ± 0.4 (P < 0.05, paired Student's t‐test), confirming that extracellular calcium is necessary for rescue of cholinergic SK responses. Together, these results support the hypothesis that cholinergic signalling rapidly depletes intracellular calcium stores at RMPs, but that, when paired with subthreshold depolarization, mAChRs also facilitate rapid refilling of stores by promoting the influx of extracellular calcium.

Figure 7. Recovery of cholinergic response rundown requires extracellular calcium.

Figure 7

Open in a new tab

A, examples of responses from two neurons showing ‘rundown’ of cholinergic SK responses during three applications of ACh (grey arrows), followed by test applications of ACh (black arrows) paired with subthreshold current steps in the presence of focally applied normal aCSF (black traces) or calcium‐free aCSF (3 mm Mg2+; red traces). The pressure of calcium‐free aCSF delivery was released right before current injections to reduce the probability of triggering action potentials. Recovery of SK responses was blocked fully (top traces), or greatly reduced (bottom traces), by local perfusion with calcium‐free aCSF. RMPs were −72 mV (top traces) and −80 mV (bottom traces). B, comparison of initial and test SK response amplitudes in baseline conditions (black bars) or after local perfusion with calcium‐free aCSF (red bars). Trials in baseline conditions involved either local perfusion with normal aCSF (n = 6), or no local perfusion (n = 5). * P < 0.05.

Discussion

ACh generates two distinct physiological responses in pyramidal neurons, including transient inhibition mediated by SK‐type calcium‐activated potassium channels (Gulledge & Stuart, 2005), and longer‐lasting excitation involving one or more non‐specific cation conductances (Andrade, 1991; Haj‐Dahmane & Andrade, 1996, 1998; Klink & Alonso, 1997; Egorov et al., 2002, 2003; Shalinsky et al., 2002; Magistretti et al. 2004; Yan et al. 2009; Zhang & Seguela, 2010; Rahman & Berger, 2011; Zhang et al. 2011; Dasari et al. 2013; Lei et al. 2014). Paradoxically, these opposing actions of ACh on pyramidal neuron excitability are mediated by a single mAChR subtype, the Gq‐coupled M1 receptor (Gulledge et al. 2009; Dasari & Gulledge, 2011). In this study we explored the rundown and rescue of cholinergic SK responses at subthreshold membrane potentials, testing the ‘unifying hypothesis’ that cholinergic inhibition and excitation in cortical pyramidal neurons represent complementary mechanisms governing calcium cycling between the endoplasmic reticulum, the cytosol and the extracellular space (Fig. 8). According to this model, M1 signal transduction stimulates calcium release from intracellular stores via IP3‐dependent calcium release, while also facilitating voltage‐dependent calcium influx to refill intracellular calcium stores. Our results suggest that store refilling occurs rapidly, within ∼200 ms, at subthreshold membrane potentials, and is sufficient to rescue cholinergic SK responses generated by already‐in‐progress mAChR‐mediated signal transduction.

Figure 8. Model of cholinergic regulation of calcium cycling in cortical neurons.

Figure 8

Open in a new tab

ACh acts primarily through Gq‐coupled M1 subtype muscarinic receptors (Gulledge et al. 2009) that act via phospholipase C (PLC) to generate inositol trisphosphate (IP3), thereby triggering calcium release from IP3 receptors in the endoplasmic reticulum (ER). Released calcium activates SK‐type potassium channels (Gulledge & Stuart, 2005), but fails to return to the ER and is instead extruded from the cytosol via the plasma membrane calcium ATPase (PMCA) and sodium‐calcium exchanger (NCX; Proulx et al. 2014). M1 activation also promotes excitation, and refilling of calcium stores, through an as‐yet‐unknown cadmium‐ and voltage‐sensitive calcium‐permeable conductance.

Brief periods of action potential generation prime cortical pyramidal neurons for Gq‐mediated calcium‐release from intracellular stores, allowing for robust SK responses that otherwise rundown rapidly in the absence of depolarization (Gulledge & Stuart, 2005; Gulledge et al. 2007). Two alternative, but not mutually exclusive, hypotheses may explain this behaviour. In one scenario, mAChR signal transduction is voltage‐dependent and enhanced by depolarization. Indeed, mAChRs exhibit modest voltage sensitivity, and gating currents for both M1 and M2 receptors have been measured (Ben‐Chaim et al. 2003, 2006; Rinne et al. 2015). However, several observations suggest that voltage‐dependency of cholinergic single transduction does not contribute to SK response rundown at RMPs. First, when neurons are ‘primed’ with suprathreshold current injections, subsequent SK responses occur reliably at RMPs and are associated with calcium transients of similar magnitude to those occurring during sustained suprathreshold depolarization (see Fig. 1). Second, following rundown of cholinergic inhibition, test applications of ACh at RMPs can generate small depolarizing responses (for example, see Fig. 5 A), suggesting that at least some aspects of mAChR signal transduction occur at RMPs in the absence of priming. Third, delaying subthreshold depolarizing steps by up to 500 ms after ACh applications has little impact on SK response latency (see Fig. 6), a finding inconsistent with voltage‐dependent mAChR signal transduction. Finally, subthreshold depolarizations that rescue cholinergic inhibition are small (∼8 mV), while the reported voltage‐sensitivity of M1 receptors is extremely weak, with already robust responses at −90 mV being enhanced only by ∼0.5% per millivolt of depolarization from −90 to +60 mV (Rinne et al. 2015). Thus, voltage sensitivity of mAChRs is unlikely to explain rundown of cholinergic SK responses at RMPs.

A second hypothesis, that the rundown of SK responses reflects depletion of intracellular calcium stores that are subsequently refilled by pairing mAChR activation with subthreshold depolarization, is supported by our findings that SK responses are time‐locked to ACh delivery, rather than to the timing of depolarization (see Fig. 6), and that SK response recovery requires the presence of extracellular calcium (see Fig. 7). However, such calcium store refilling would need to be rapid enough to allow calcium flow through already open IP3 receptors to trigger SK responses, as Gq‐triggered SK responses are absent when sarco/endoplasmic reticulum calcium‐ATPase (SERCA) pumps are blocked (Fiorillo & Williams, 1998; Gulledge & Stuart, 2005; Gulledge & Kawaguchi, 2007; Power & Sah, 2007; Hagenston et al. 2008; Power & Sah, 2008; El‐Hassar et al. 2011). Can calcium store refilling occur rapidly enough (within ∼100 ms; see Fig. 6) to rescue SK responses? In cardiomyocytes, calcium cycling through intracellular stores can occur extremely rapidly. For instance, some small birds and mammals have resting heart rates approaching 600 beats per minute (∼10 Hz; Detweiler & Erickson, 2004), and hummingbird heart rates can reach 1200 beats per minute (∼20 Hz) during flight (Lasiewski, 1964). Thus, calcium cycling into and out of the sarcoplasmic reticulum of some cardiomyocytes occurs at 50–100 ms intervals, suggesting it is physiologically feasible that mAChR‐stimulated calcium influx and SERCA‐mediated store refilling occur rapidly enough to rescue previously rundown SK responses.

Which ion channels mediate calcium influx during mAChR stimulation? A variety of ionic effectors probably contribute to cholinergic excitation of cortical neurons (Andrade, 1991; Haj‐Dahmane & Andrade, 1996, 1998; Yan et al. 2009; Zhang et al. 2011; Dasari et al. 2013; Thuault et al. 2013; Lei et al. 2014; Kurowski et al. 2015). Some of these effectors, including transient receptor potential cation (TRPC) (Yan et al. 2009; Zhang et al. 2011) channels, are attractive candidates because they are both calcium permeable and gated by Gq signal transduction (Wu et al. 2010). However, both cholinergic excitation and rescue of SK responses are voltage‐sensitive, and facilitated by depolarization above RMPs. While homomeric TRPC channels typically lack this voltage sensitivity, heteromeric TRPC channels comprising TRPC1 and TRPC5 subunits exhibit negative slope conductances at subthreshold membrane potentials (Strübing et al. 2001). The diversity of potential heteromeric TRP channels may explain the robustness of cholinergic excitation across neuron types and animal species, as well as the range of results from pharmacological and gene knockout studies aiming to identify the ionic effectors mediating cholinergic excitation (Yan et al. 2009; Zhang et al. 2011; Dasari et al. 2013; Lei et al. 2014). Alternatively, the voltage dependency of SK response rescue may be indirect, as calcium entry through T‐type (Yan et al. 2009) or L‐type (Power & Sah, 2005) calcium channels at subthreshold membrane potentials may facilitate or gate Gq‐triggered activation of TRP‐like channels. Additional studies will be necessary to confirm whether TRPC1/TRPC5, or other heteromeric TRP channels, alone or in combination with voltage‐gated calcium channels, are responsible for the store‐refilling engaged by M1 receptors in cortical pyramidal neurons.

Functional relevance of muscarinic signalling in pyramidal neurons

Phasic release of ACh in the prefrontal cortex facilitates a range of cognitive processes, including attentiveness to salient environmental cues (Parikh et al. 2007; Gritton et al. 2016). While cholinergic responses in pyramidal neurons exhibit a degree location‐ and layer‐dependent specificity (Gulledge et al. 2007; Dembrow et al. 2010; Hedrick & Waters, 2015), the signalling mechanisms engaged by mAChRs, including transient SK‐driven inhibition and longer‐lasting excitation, are robust and generally conserved in pyramidal cells across the neocortex and in the hippocampus.

How do these two opposing responses, inhibition and excitation, contribute to cortical processing? At any given moment, fluctuations in the membrane potentials of individual pyramidal neurons within a population reflect unique patterns of ongoing synaptic input. Some neurons will be close to, or at, action potential threshold, while others may be hyperpolarized and quiescent. We propose that transient ACh release in the cortex, as occurs in response to salient environmental cues (Parikh et al. 2007), may help focus attention on relevant sensory input by transiently driving populations of pyramidal neurons toward a common hyperpolarized membrane potential via M1‐receptor‐driven calcium release from intracellular stores and SK channel activation. Such hyperpolarization may serve to ‘reset’ synaptic integration across populations of pyramidal neurons, such that future action potential generation more accurately reflects updated, rather than previous, patterns of afferent input. At the same time, voltage‐dependent cholinergic excitation will preferentially enhance the action potential output of neurons that subsequently receive threshold levels of new excitatory drive, while neurons receiving less drive will remain quiescent. In this way, ACh may selectively enhance the output of neurons most relevant to perceiving, or generating behavioural responses to, the stimuli triggering ACh release. Finally, by enhancing voltage‐dependent calcium influx, cholinergic excitation effectively refills the intracellular calcium stores responsible for SK‐mediated inhibition, therefore ensuring inhibitory signalling fidelity in response to future ACh transients.

Additional information

Competing interests

The authors declare that they have no conflict of interest.

Author contributions

ATG and SD conceived the project and designed the experiments. SD, CH and ATG contributed to data collection, analysis and interpretation. ATG wrote the manuscript, which was approved by all authors. All persons designated as authors qualify for authorship, and all those who qualify for authorship are listed.

Funding

This study was supported by NIMH grants R01 MH083806 and R01 MH099054.

Acknowledgements

The authors thank Saiko Ikeda for technical assistance and Arielle L. Baker for comments on the manuscript.

This is an Editor's Choice article from the 1 March 2017 issue.

References

  1. Andrade R (1991). Cell excitation enhances muscarinic cholinergic responses in rat association cortex. Brain Res 548, 81–93. [DOI] [PubMed] [Google Scholar]
  2. Ben‐Chaim Y, Chanda B, Dascal N, Bezanilla F, Parnas I & Parnas H (2006). Movement of ‘gating charge’ is coupled to ligand binding in a G‐protein‐coupled receptor. Nature 444, 106–109. [DOI] [PubMed] [Google Scholar]
  3. Ben‐Chaim Y, Tour O, Dascal N, Parnas I & Parnas H (2003). The M2 muscarinic G‐protein‐coupled receptor is voltage‐sensitive. J Biol Chem 278, 22482–22491. [DOI] [PubMed] [Google Scholar]
  4. Bezprozvanny I, Watras J & Ehrlich BE (1991). Bell‐shaped calcium‐response curves of Ins(1,4,5)P3‐ and calcium‐gated channels from endoplasmic reticulum of cerebellum. Nature 351, 751–754. [DOI] [PubMed] [Google Scholar]
  5. Dasari S, Abramowitz J, Birnbaumer L & Gulledge AT (2013). Do canonical transient receptor potential channels mediate cholinergic excitation of cortical pyramidal neurons? Neuroreport 24, 550–554. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Dasari S & Gulledge AT (2011). M1 and M4 receptors modulate hippocampal pyramidal neurons. J Neurophysiol 105, 779–792. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Dembrow NC, Chitwood RA & Johnston D (2010). Projection‐specific neuromodulation of medial prefrontal cortex neurons. J Neurosci 30, 16922–16937. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Detweiler D & Erickson H (2004). Regulation of the heart In Dukes’ Physiology of Domestic Animals, ed. Reece W, pp. 261–274. Cornell University Press, Ithaca, NY. [Google Scholar]
  9. Egorov AV, Angelova PR, Heinemann U & Müller W (2003). Ca2+‐independent muscarinic excitation of rat medial entorhinal cortex layer V neurons. Eur J Neurosci 18, 3343–3351. [DOI] [PubMed] [Google Scholar]
  10. Egorov AV, Hamam BN, Fransen E, Hasselmo ME & Alonso AA (2002). Graded persistent activity in entorhinal cortex neurons. Nature 420, 173–178. [DOI] [PubMed] [Google Scholar]
  11. El‐Hassar L, Hagenston AM, D'Angelo LB & Yeckel MF (2011). Metabotropic glutamate receptors regulate hippocampal CA1 pyramidal neuron excitability via Ca2+ wave‐dependent activation of SK and TRPC channels. J Physiol 589, 3211–3229. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Fiorillo CD & Williams JT (1998). Glutamate mediates an inhibitory postsynaptic potential in dopamine neurons. Nature 394, 78–82. [DOI] [PubMed] [Google Scholar]
  13. Gritton HJ, Howe WM, Mallory CS, Hetrick VL, Berke JD & Sarter M (2016). Cortical cholinergic signalling controls the detection of cues. Proc Natl Acad Sci USA 113, E1089–1097. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Gulledge AT, Bucci DJ, Zhang SS, Matsui M & Yeh HH (2009). M1 receptors mediate cholinergic modulation of excitability in neocortical pyramidal neurons. J Neurosci 29, 9888–9902. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Gulledge AT & Kawaguchi Y (2007). Phasic cholinergic signalling in the hippocampus: functional homology with the neocortex? Hippocampus 17, 327–332. [DOI] [PubMed] [Google Scholar]
  16. Gulledge AT, Park SB, Kawaguchi Y & Stuart GJ (2007). Heterogeneity of phasic cholinergic signalling in neocortical neurons. J Neurophysiol 97, 2215–2229. [DOI] [PubMed] [Google Scholar]
  17. Gulledge AT & Stuart GJ (2003). Excitatory actions of GABA in the cortex. Neuron 37, 299–309. [DOI] [PubMed] [Google Scholar]
  18. Gulledge AT & Stuart GJ (2005). Cholinergic inhibition of neocortical pyramidal neurons. J Neurosci 25, 10308–10320. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Hagenston AM, Fitzpatrick JS & Yeckel MF (2008). MGluR‐mediated calcium waves that invade the soma regulate firing in layer V medial prefrontal cortical pyramidal neurons. Cereb Cortex 18, 407–423. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Haj‐Dahmane S & Andrade R (1996). Muscarinic activation of a voltage‐dependent cation nonselective current in rat association cortex. J Neurosci 16, 3848–3861. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Haj‐Dahmane S & Andrade R (1998). Ionic mechanism of the slow afterdepolarization induced by muscarinic receptor activation in rat prefrontal cortex. J Neurophysiol 80, 1197–1210. [DOI] [PubMed] [Google Scholar]
  22. Hedrick T & Waters J (2015). Acetylcholine excites neocortical pyramidal neurons via nicotinic receptors. J Neurophysiol 113, 2195–2209. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Klink R & Alonso A (1997). Ionic mechanisms of muscarinic depolarization in entorhinal cortex layer II neurons. J Neurophysiol 77, 1829–1843. [DOI] [PubMed] [Google Scholar]
  24. Kurowski P, Gawlak M & Szulczyk P (2015). Muscarinic receptor control of pyramidal neuron membrane potential in the medial prefrontal cortex (mPFC) in rats. Neuroscience 303, 474–488. [DOI] [PubMed] [Google Scholar]
  25. Lasiewski RC (1964). Body temperatures, heart and breathing rate, and evaporative water loss in hummingbirds. Physiol Zool 37, 212–223. [Google Scholar]
  26. Lei YT, Thuault SJ, Launay P, Margolskee RF, Kandel ER & Siegelbaum SA (2014). Differential contribution of TRPM4 and TRPM5 nonselective cation channels to the slow afterdepolarization in mouse prefrontal cortex neurons. Front Cell Neurosci 8, 267. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Magistretti J, Ma L, Shalinsky MH, Lin W, Klink R & Alonso A (2004). Spike patterning by Ca2+‐dependent regulation of a muscarinic cation current in entorhinal cortex layer II neurons. J Neurophysiol 92, 1644–1657. [DOI] [PubMed] [Google Scholar]
  28. Parikh V, Kozak R, Martinez V & Sarter M (2007). Prefrontal acetylcholine release controls cue detection on multiple timescales. Neuron 56, 141–154. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Power JM & Sah P (2005). Intracellular calcium store filling by an L‐type calcium current in the basolateral amygdala at subthreshold membrane potentials. J Physiol 562, 439–453. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Power JM & Sah P (2007). Distribution of IP3‐mediated calcium responses and their role in nuclear signalling in rat basolateral amygdala neurons. J Physiol 580, 835–857. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Power JM & Sah P (2008). Competition between calcium‐activated K+ channels determines cholinergic action on firing properties of basolateral amygdala projection neurons. J Neurosci 28, 3209–3220. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Proulx E, Suri D, Heximer SP, Vaidya VA & Lambe EK (2014). Early stress prevents the potentiation of muscarinic excitation by calcium release in adult prefrontal cortex. Biol Psychiatry 76, 315–323. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Rahman J & Berger T (2011). Persistent activity in layer 5 pyramidal neurons following cholinergic activation of mouse primary cortices. Eur J Neurosci 34, 22–30. [DOI] [PubMed] [Google Scholar]
  34. Rinne A, Mobarec JC, Mahaut‐Smith M, Kolb P & Bunemann M (2015). The mode of agonist binding to a G protein‐coupled receptor switches the effect that voltage changes have on signalling. Sci Signal 8, ra110. [DOI] [PubMed] [Google Scholar]
  35. Shalinsky MH, Magistretti J, Ma L & Alonso AA (2002). Muscarinic activation of a cation current and associated current noise in entorhinal‐cortex layer‐II neurons. J Neurophysiol 88, 1197–1211. [DOI] [PubMed] [Google Scholar]
  36. Strübing C, Krapivinsky G, Krapivinsky L & Clapham DE (2001). TRPC1 and TRPC5 form a novel cation channel in mammalian brain. Neuron 29, 645–655. [DOI] [PubMed] [Google Scholar]
  37. Thuault SJ, Malleret G, Constantinople CM, Nicholls R, Chen I, Zhu J, Panteleyev A, Vronskaya S, Nolan MF, Bruno R, Siegelbaum SA & Kandel ER (2013). Prefrontal cortex HCN1 channels enable intrinsic persistent neural firing and executive memory function. J Neurosci 33, 13583–13599. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Wu LJ, Sweet TB & Clapham DE (2010). International Union of Basic and Clinical Pharmacology. LXXVI. Current progress in the mammalian TRP ion channel family. Pharmacol Rev 62, 381–404. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Yan HD, Villalobos C & Andrade R (2009). TRPC channels mediate a muscarinic receptor‐induced afterdepolarization in cerebral cortex. J Neurosci 29, 10038–10046. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Zhang Z, Reboreda A, Alonso A, Barker PA & Seguela P (2011). TRPC channels underlie cholinergic plateau potentials and persistent activity in entorhinal cortex. Hippocampus 21, 386–397. [DOI] [PubMed] [Google Scholar]
  41. Zhang Z & Seguela P (2010). Metabotropic induction of persistent activity in layers II/III of anterior cingulate cortex. Cereb Cortex 20, 2948–2957. [DOI] [PubMed] [Google Scholar]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society

RESOURCES