Facilitation of long-term potentiation by muscarinic M(1) receptors is mediated by inhibition of SK channels

Abstract

Muscarinic receptor activation facilitates the induction of synaptic plasticity and enhances cognitive function. However, the specific muscarinic receptor subtype involved and the critical intracellular signaling pathways engaged have remained controversial. Here, we show that the recently discovered highly selective allosteric M(1) receptor agonist 77-LH-28-1 facilitates long-term potentiation (LTP) induced by theta burst stimulation at Schaffer collateral synapses in the hippocampus. Similarly, release of acetylcholine by stimulation of cholinergic fibers facilitates LTP via activation of M(1) receptors. N-methyl-D-aspartate receptor (NMDAR) opening during theta burst stimulation was enhanced by M(1) receptor activation, indicating this is the mechanism for LTP facilitation. M(1) receptors were found to enhance NMDAR activation by inhibiting SK channels that otherwise act to hyperpolarize postsynaptic spines and inhibit NMDAR opening. Thus, we describe a mechanism where M(1) receptor activation inhibits SK channels, allowing enhanced NMDAR activity and leading to a facilitation of LTP induction in the hippocampus.

Figure 1

The Effect of Cholinergic Agonists on the Cellular and Synaptic Properties of CA1 Pyramidal Cells (A) Bath application of 1 μM carbachol caused a depolarization of the membrane potential (middle) and an increase in input resistance (bottom). Top: Sample voltage traces from a single experiment demonstrating the voltage response to a 100 pA current step during baseline (1) and in the presence of 1 μM carbachol (2). The dotted line represents the baseline membrane potential and the overlay shows the offset traces. (B) Bath application of 10 μM 77-LH-28-1 caused a depolarization of the membrane potential (middle) and increase in input resistance (bottom) that reversed on washout. Top: Sample voltage traces as described in (A). (C) The effect of 10 μM 77-LH-28-1 on the membrane potential (middle) and input resistance (bottom) of CA1 pyramidal cells was blocked by the coapplication of 25 μM pirenzepine (black bar). Top: Sample traces as described in (A). The scale bars represent 2 mV, 100 ms. Summary bar graph of the concentration-dependent membrane potential (D) or input resistance (E) changes 8–10 min after application of cholinergic agonists and antagonists. ^∗ indicates a significant difference from baseline values in (D) and (E) (p < 0.05). Bath application of 10 μM 77-LH-28-1 caused a depolarization of the membrane potential (F) and an increase in input resistance (G) in slices taken from M₁+/+ mice but not M₁−/− mice. (H) Bath application of 1 μM carbachol caused a depression of the evoked EPSC amplitude. Right: Sample EPSC traces from a single experiment during baseline and in the presence of 1 μM carbachol. The scale bars represent 10 pA, 20 ms. (I) Bath application of 10 μM 77-LH-28-1 resulted in no change in EPSC amplitude. Right: Sample EPSC traces as described in (H). The scale bars represent 20 pA, 20 ms. The data are plotted as the mean ± SEM.

Figure 2

The M₁ Receptor Agonist 77-LH-28-1 Facilitates the Induction of LTP by TBP (A) Diagram of TBP protocol. Top: Voltage trace of TBP protocol recorded at the soma of a CA1 pyramidal cell. The scale bars represent 40 mV, 100 ms. Middle and bottom: Traces illustrate the timing of inputs to the stimulating and recording electrodes evoking EPSPs and somatic action potentials, respectively. (B) TBP does not induce LTP under control conditions. Bottom: Coincident TBP of subthreshold EPSPs and somatic action potentials induced no change in EPSC amplitude in the test (black circles) or control (white circles) pathways. The arrow indicates the timing of the TBP protocol. Top: Example of voltage traces from a single experiment showing the initial burst of five coincident EPSPs and action potentials (black) and a single test burst of five subthreshold EPSPs (gray). The scale bars represent 20 mV, 20 ms. Middle: Example of current traces from a single experiment illustrating the mean EPSC response during the baseline (1) and at 30–35 min (2) in the test and control pathways. The scale bars represent 10 pA, 40 ms. (C) TBP does induce LTP in the presence of the M₁ receptor agonist 77-LH-28-1. Bottom: In the presence of 77-LH-28-1 (10 μM), coincident TBP of subthreshold EPSPs and somatic action potentials induced pathway-specific LTP. Symbols as described in (B). Top: Example of voltage traces as described in (B). The scale bars represent 20 mV, 20 ms. Middle: Example of EPSC current traces from a single experiment as described in (B). The scale bars represent 20 pA, 40 ms. (D) TBP does not induce LTP in the presence of the M₁ receptor agonist 77-LH-28-1 and the M₁ receptor antagonist pirenzepine. Bottom: Coapplication of pirenzepine (25 μM) and 77-LH-28-1 (10 μM) prevented the induction of pathway-specific LTP. Symbols as described in (B). Top: Example of voltage traces as described in (B). The scale bars represent 20 mV, 20 ms. Middle: Example of EPSC current traces from a single experiment as described in (B). The scale bars represent 10 pA, 40 ms. The data are plotted as the mean ± SEM.

Figure 3

The Endogenous Release of Acetylcholine Acting at M₁ Receptors Facilitates the Induction of LTP by TBP (A) mAChR-mediated EPSPs can be induced in hippocampal slices. Top left: Schematic of a hippocampal slice demonstrating the positioning of the stimulating electrodes in stratum radiatum and stratum oriens (s.o.). Top right: Example of experiment shows the slow EPSP evoked by stratum oriens stimulation was blocked by the M₁ receptor antagonist pirenzepine (25 μM). Bottom: Example of voltage traces of the slow EPSP in control conditions and in the presence of pirenzepine (25 μM). The scale bars represent 10 mV, 2 s. (B) Illustration of mAChR-mediated EPSP and TBP protocol. Top: Example of voltage trace of the TBP protocol recorded at the soma during stimulation of the slow mAChR-mediated EPSP. The scale bars represent 2 mV, 2 s. Bottom Left: Schematic of the initial train of the TBP protocol illustrating the timing of the TBP in relation to stratum oriens stimulation. The scale bars represent 20 mV, 200 ms. Bottom right: Example of voltage trace from a single experiment showing the initial burst of five coincident EPSPs and action potentials (black) and a single test burst of five subthreshold EPSPs (gray) applied before stratum oriens stimulation. The scale bars represent 20 mV, 20 ms. (C) TBP does induce test pathway-specific LTP with simultaneous stimulation of the muscarinic EPSP. Arrow indicates the timing of the concurrent stratum oriens stimulation and TBP. Top: Example of EPSC current traces from a single experiment illustrating the mean EPSC response during the baseline (1) and at 30–35 min (2) in the test (black circles) and control (white circles) pathways. The scale bars represent 20 pA, 40 ms. (D) The M₁ receptor antagonist pirenzepine (25 μM) prevented the induction of LTP by concurrent stratum oriens stimulation and TBP. Symbols as described in (C). Top: Example of EPSC current traces from a single experiment as described in (C). The scale bars represent 10 pA, 40 ms. The data are plotted as the mean ± SEM.

Figure 4

77-LH-28-1 Does Not Alter the Amplitude or Decay of Isolated NMDA EPSCs (A) Bath application of 77-LH-28-1 (10 μM) did not alter the amplitude of isolated NMDA receptor-mediated EPSCs, which were completely blocked by the bath application of D-AP5 (50 μM). Right: Example of current traces illustrating the mean EPSC response during the baseline (1), during the application of 77-LH-28-1 (2), and in the presence of D-AP5 (3). The scale bars represent 40 pA, 100 ms. (B) Bar graph showing the mean decay time constant of the NMDA receptor-mediated EPSC during the baseline and in the presence of 77-LH-28-1. The bar graph is overlaid with data from individual experiments. (C) Bath application of carbachol (10 μM) increased NMDA responses to exogenous NMDA application (1 mM) at −60 mV but not at +40 mV holding potential. Right: Example of traces of NMDA responses before (1) and after (2) carbachol application at the two membrane potentials. The scale bars represent 50 pA, 2 s. The data are plotted as the mean ± SEM.

Figure 5

M₁ Receptor Activation Enhances the NMDAR-Mediated Component of EPSPs (A) 77-LH-28-1 (10 μM) prolonged the duration of single EPSPs. Left: Example of voltage traces in the presence of 77-LH-28-1 (red) or control conditions (black). The scale bars represent 0.6 mV, 40 ms. Right: Average decay time constant increase. (B) 77-LH-28-1 (10 μM) prolonged the membrane decay time constant in response to a short subthreshold current injection. Left: Example of voltage traces in the presence of 77-LH-28-1 (red) or control conditions (black). The scale bars represent 0.6 mV, 20 ms. Right: Average membrane decay time constant increase. (C) Summated EPSPs during synaptic theta burst stimulation do not exhibit an NMDAR-mediated component. Left: Example of voltage traces of a burst of five EPSPs under control conditions (black) and in the presence of 50 μM D-AP5 (green). The scale bars represent 1.5 mV, 40 ms. Right: The average normalized decay time constant for a burst of five EPSPs does not change in the presence of D-AP5. (D) 77-LH-28-1 enabled NMDAR activation during synaptic theta burst stimulation. Left: Example of voltage traces showing a burst of five EPSPs under control conditions (black), in the presence of 10 μM 77-LH-28-1 (red), and after addition of 50 μM D-AP5 (green). The scale bars represent 2 mV, 40 ms. Right: The average normalized decay time constant is significantly reduced by addition of D-AP5 in the presence of 77-LH-28-1. (E) 77-LH-28-1 had no effect on EPSP duration in the presence of pirenzepine. Left: Example of voltage traces showing a burst of five EPSPs under control conditions (black) and in the presence of 10 μM 77-LH-28-1 (red) all in the presence of pirenzepine (25 μM). The scale bars represent 2 mV, 40 ms. Right: The average normalized decay time constant is not increased by 77-LH-28-1 in the presence of pirenzepine. (F) Oxotremorine-m enabled NMDAR activation during synaptic theta burst stimulation. Left: Example of voltage traces showing a burst of five EPSPs under control conditions (black), in the presence of 10 μM oxotremorine-m (red), and after addition of 50 μM D-AP5 (green). The scale bars represent 2 mV, 40 ms. Right: The average normalized decay time constant is significantly reduced by addition of D-AP5 in the presence of oxotremorine-m (oxo-m). (G) 77-LH-28-1 had no effect on EPSP duration in slices taken from M₁−/− mice. Left: Example of voltage traces showing a burst of five EPSPs under control conditions (black) and in the presence of 10 μM 77-LH-28-1 (red) in M₁+/+ and M₁−/− mice. The scale bars represent 2 mV, 40 ms. Right: The average normalized decay time constant is increased in the presence of 77-LH-28-1 in M₁+/+ mice but not M₁−/− mice. ^∗ indicates significant difference (p < 0.05). The data are plotted as the mean ± SEM.

Figure 6

The M₁ Receptor Agonist 77-LH-28-1 Prolongs the NMDAR-Mediated Component of EPSPs via Inhibition of SK Channels (A) Apamin prolongs the duration of summated EPSPs and occludes the action of 77-LH-28-1. Example of peak normalized voltage traces showing a burst of five EPSPs under control conditions (black), in the presence of 100 nM apamin (red), and after addition of 10 μM 77-LH-28-1 (green). The scale bar represents 40 ms. The average normalized decay time constant is significantly prolonged by addition of apamin with no further change in the presence of 77-LH-28-1. (B) Prolongation of EPSPs by apamin is reversed by application of D-AP5. Example of peak normalized voltage traces showing a burst of five EPSPs under control conditions (black), in the presence of 100 nM apamin (red), and after addition of 50 μM D-AP5 (green). The scale bar represents 40 ms. The average normalized decay time constant is significantly prolonged by addition of apamin and reversed by D-AP5. (C) The KCNQ channel blocker XE-991 prolongs the duration of summated EPSPs but does not occlude the action of 77-LH-28-1. Example of peak normalized voltage traces showing a burst of five EPSPs under control conditions (black), in the presence of 10 μM XE-991 (red), and after addition of 10 μM 77-LH-28-1 (green). The scale bar represents 40 ms. The average normalized decay time constant is significantly prolonged by addition of XE-991, but there is an additional significant prolongation with 77-LH-28-1. (D) Prolongation of EPSPs by XE-991 is partially reversed by application of D-AP5. Example of peak normalized voltage traces showing a burst of five EPSPs under control conditions (black), in the presence of 10 μM XE-991 (red), and after addition of 50 μM D-AP5 (green). The scale bar represents 40 ms. The average normalized decay time constant is significantly prolonged by addition of XE-991 and partially reversed by D-AP5. (E) Apamin prolongs the duration of summated EPSPs but does not occlude the action of XE-991. Example of peak normalized voltage traces showing a burst of five EPSPs under control conditions (black), in the presence of 100 nM apamin (red), and after addition of 10 μM XE-991 (green). The scale bar represents 40 ms. The average normalized decay time constant is significantly prolonged by addition of apamin, but there is additional significant prolongation with XE-991. (F) I_AHPs recorded from CA1 pyramidal cells in the perforated patch configuration are primarily composed of current through SK channels. I_AHPs are stimulated in the presence of TTX (1 μM) by switching the membrane potential from −50 mV to +10 mV for 100 ms. The I_AHP is seen after return to −50 mV. Application of XE-991 (10 μM) reduced I_AHP by 10% ± 2%; apamin (100 nM) blocked the remainder. The scale bars represent 50 pA, 50 ms. (G) 77-LH-28-1 (10 μM) inhibited the SK channel-mediated component of the I_AHP that partially recovered on washout. Subtraction of the apamin-insensitive component of the I_AHP revealed the SK-channel-mediated component (right). The scale bars represent 50 pA, 50 ms. (H) 77-LH-28-1 had no effect on I_AHP after application of 100 nM apamin. The scale bars represent 75 pA, 50 ms. (I) 77-LH-28-1 had no effect on I_AHP after incubation in 25 μM pirenzepine. The scale bars represent 30 pA, 50 ms. (J) Oxotremorine-m (oxo-m, 10 μM) inhibited the SK channel-mediated component of the I_AHP. The scale bars represent 30 pA, 50 ms. (K) Summary of the effects of M₁ receptor activation on I_AHP measured by pharmacological subtraction after apamin application. ^∗ denotes statistical significance (p < 0.05). The data are plotted as the mean ± SEM.

Figure 7

PKC Mediates the M₁ Receptor-Induced Inhibition of SK Channels (A) Inhibition of I_AHP by 77-LH-28-1 (10 μM) is reduced by preincubation with the PKC inhibitor Go6976 (100 nM). The scale bars represent 30 pA, 50 ms. (B) Prolongation of EPSPs by 77-LH-28-1 is blocked by preincubation with the PKC inhibitor Go6976 (100 nM). Example of peak normalized voltage traces showing a burst of five EPSPs after preincubation with Go6976 (black) and no prolongation in the presence of 10 μM 77-LH-28-1 (red). The scale bar represents 40 ms. The average normalized decay time constant is not significantly prolonged. (C) Prolongation of EPSPs by 77-LH-28-1 is blocked by inclusion of the PKC inhibitor PKC 19-36 in the patch pipette. Example of peak normalized voltage traces showing a burst of five EPSPs after infusion of (Glu27)PKC 19-36 or PKC 19-36 (black). Prolongation in the presence of 10 μM 77-LH-28-1 (red) only occurs in the presence of (Glu27)PKC 19-36. The scale bar represents 40 ms. The average normalized decay time constant is significantly prolonged by 77-LH-28-1 in the presence of (Glu27)PKC 19-36 and this is significantly reduced by PKC 19-36. (D) Inhibition of I_AHP by 77-LH-28-1 (10 μM) is unchanged by preincubation with the CK2 inhibitors TBB (10 μM) or TMCB (10 μM). The scale bars represent 30 pA, 50 ms. (E) Incubation in TMCB (10 μM) or TBB (10 μM) greatly prolonged the EPSP compared to control. Example of peak normalized voltage traces showing a burst of five EPSPs under control conditions (black) and, in separate experiments, after incubation in 10 μM TMCB (red) or 10 μM TBB (green). The scale bar represents 60 ms. The average decay time constant is significantly prolonged by incubation in TMCB or TBB. Input resistance after incubation in TMCB or TBB was unchanged compared to control. (F) Acute application of TMCB (10 μM) or TBB (10 μM) prolonged the average normalized decay time constant of the EPSP. (G) Apamin prolongs the duration of summated EPSPs but does not occlude the action of TBB. Example of peak normalized voltage traces showing a burst of five EPSPs under control conditions (black), in the presence of 100 nM apamin (red), and after addition of 10 μM TBB (green). The scale bar represents 40 ms. The average normalized decay time constant is significantly prolonged by addition of apamin, but there is an additional significant prolongation with 30 min application of TBB. (H) Incubation in TMCB does not block the action of 77-LH-28-1. Example of peak normalized voltage traces showing a burst of five EPSPs after incubation in TMCB (black) and with addition of 10 μM 77-LH-28-1 (red). The scale bar represents 60 ms. The average normalized decay time constant shows a significant change in the presence of 77-LH-28-1. (I) Incubation in TBB blocks the action of 77-LH-28-1. Example of peak normalized voltage traces showing a burst of five EPSPs after incubation in TBB (black) and after addition of 10 μM 77-LH-28-1 (red). The scale bar represents 60 ms. The average normalized decay time constant shows no change in the presence of 77-LH-28-1. The data are plotted as the mean ± SEM.

Figure 8

SK Channel Inhibition Facilitates LTP Induction (A) SK channel blockade facilitates LTP induction. In the continuous presence of apamin (100 nM) TBP induces pathway-specific LTP. The arrow indicates the timing of the TBP protocol. Top: Example of voltage traces from a single experiment showing the initial burst of five coincident EPSPs and action potentials (black) and a single test burst of five subthreshold EPSPs (gray). The scale bars represent 25 mV, 20 ms. Middle: Example of current traces from a single experiment illustrating the mean EPSC response during the baseline (1) and at 30–35 min (2) in the test (black circles) and control (white circles) pathways. The scale bars represent 50 pA, 40 ms. (B) In the continuous presence of apamin (100 nM) and 77-LH-28-1 (10 μM), TBP induces pathway-specific LTP similar in magnitude to apamin or 77-LH-28-1 in isolation. The arrow indicates the timing of the TBP protocol. Symbols as described in (A). Top: Example of voltage traces. The scale bars represent 25 mV, 20 ms. Middle: Example of EPSC current traces from a single experiment. The scale bars represent 50 pA, 40 ms. (C) 77-LH-28-1 (10 μM), apamin (100 nM), or a combination of 77-LH-28-1 and apamin all induce a similar amount of LTP. The data are plotted as the mean ± SEM.