| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
New York University Center for Neural Science, New York, New York
Submitted 10 June 2005; accepted in final form 9 January 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
) through the activity of nicotinic and muscarinic receptors. Much of the effect of acetylcholine (ACh) may arise from modulation of excitatory (glutamatergic) synaptic transmission (Hasselmo and Bower 1992; Vidal and Changeux 1993; Gil et al. 1997; Aramakis and Metherate 1998) and consequent ‘gating’ of sensory information during arousal (Steriade et al. 2001). Past in vitro studies have used whole cell recording in conjunction with extracellular stimulation in slice preparations to examine cholinergic modulation of cortical excitatory pathways. The general consensus among these studies is that ACh enhances glutamate release at thalamo-cortical synapses via nicotinic receptors while reducing glutamate release at cortico-cortical synapses via muscarinic receptors, at least within superficial, presumably thalamo-recipient layers 2–4 (Hasselmo and Bower 1992; Vidal and Changeux 1993; Gil et al. 1997; Hsieh et al. 2000). Acetylcholine's selective enhancement of sensory responses relative to cortico-cortical responses has been corroborated in vivo using natural stimuli (Oldford and Castro-Alamancos 2003).
In addition to synaptic effects, ACh and cholinergic agonists change the excitability and firing properties of cortical pyramidal neurons, primarily through the activity of muscarinic receptors (Krnjevic et al. 1971; McCormick 1992). Moreover, ACh acts on inhibitory neurons as well, both through synaptic effects on GABA release (Metherate and Ashe 1995) and through effects on cell excitability (McCormick and Prince 1986; Christophe et al. 2002) that may vary among different inhibitory neuronal classes (Kawaguchi 1997; Xiang et al. 1998; Porter et al. 1999).
The pathway specific findings summarized above are drawn largely from examination of thalamo-cortical or relatively long-range cortico-cortical connections terminating in thalamo-recipient layers (but see Kimura et al. 1999). Extracellular stimulation methods have prevented examination of the local connections (within a few hundred microns) that are prevalent throughout cortex, and little is known in particular about the cholinergic modulation of excitatory connections in the deeper layers. Neighboring layer 5 pyramidal neurons in young rat somatosensory cortex are synaptically connected with a frequency of 
10%, and about a third of these connections are bidirectional (Markram et al. 1997; compare Thomson et al. 1993; Deuchars et al. 1994). Recurrent excitation among neighboring excitatory neurons may amplify and sharpen sensory input, or it may alter sensory receptive field properties in other ways, depending on the exact connectivity (Ben-Yishai et al. 1995; Douglas et al. 1995; Chance et al. 1999; reviewed in Sompolinsky and Shapley 1997). Application of cholinergic agonists or antagonists, or manipulation of endogenous cholinergic pathways, has been shown to change the receptive field properties of somatosensory (Metherate et al. 1988) and auditory cortical neurons (Ashe et al. 1989; McKenna et al. 1989) in vivo. We speculated that modulation of local excitatory synaptic transmission by ACh might contribute to these changes. Here we used dual intracellular recording to examine unitary EPSPs evoked between pairs of layer 5 pyramidal neurons. This approach has allowed us to analyze the effects of ACh on a specific intra-cortical circuit that has not been accessible to previous studies using extracellular stimulation.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Details of the slice preparation and recording were as described previously (Reyes and Sakmann 1999; Chance et al. 2002), and conformed to protocols approved by the New York University Animal Welfare Committee. Wistar rats (postnatal day 10–26) were decapitated after anesthesia via halothane inhalation, and 300 µm parasagittal slices containing somatosensory cortex were made using a vibrating microtome (Leica Instruments GmbH, Nussloch, Germany). Slices were placed in the recording chamber and perfused with oxygenated artificial cerebrospinal fluid (ACSF: 125 mM NaCl, 25 mM NaHCO3, 25 mM glucose, 2.5 mM KCl, 1.25 mM NaH2PO4, 2 mM CaCl2, 1 mM MgCl2, all purchased from Sigma-Aldrich, St Louis, MO) at 32–34°C (flow rate 1–2 mL/min). In some experiments, standard ACSF was replaced by magnesium-free ACSF (0 mM MgCl2, 3 mM CaCl2, otherwise as described above) to unmask the activity of N-methyl-D-aspartate receptors (NMDAR). Inclusion of 40 µM glycine (NMDAR cofactor) in some experiments did not affect evoked EPSPs appreciably.
Slices were viewed under infrared differential interference contrast (IR-DIC) microscopy (Stuart et al. 1993). Layer 5 pyramidal cells were identified by their large somata and prominent apical dendrites. The somata of the recorded cells were always <350 µm apart, and usually <100 µm apart. Whole cell recordings were made from up to four cells simultaneously, using pipettes with 8–12 MOhm resistance when filled with 100 mM K-gluconate, 20 mM KCl, 4 mM ATP-Mg, 10 mM phosphocreatine, 0.3 mM GTP, and 10 mM HEPES, pH 7.3. Recordings were made using BVC-700A amplifiers (Dagan, Minneapolis, MN), digitized at 10 kHz using an ITC-18 interface (Instrutech, Port Washington, NY), and stored on a computer using Igor software (Wavemetrics, Lake Oswego, OR). Voltages were corrected for liquid junction potential using the offset control on the amplifier. Input resistance was measured on initial break-in by measuring the voltage response to 1 s incremental current steps.
For voltage clamp recording, the intracellular solution for the postsynaptic cell was as described above, except that Cs-gluconate was substituted for K-gluconate, and 5 mM QX-314 (Tocris Cookson, Ellisville, MO) was included to block voltage-dependent Na channels.
In synaptically connected cells, suprathreshold stimulation evoked unitary EPSPs in the target cell(s). Connections were monosynaptic as evidenced by short average latency, monophasic rising component, and small average amplitude (see RESULTS). Presynaptic cells were stimulated with a series of 1 to 10 pulses, each of 5 ms duration, at 10 Hz (3 pulses were used in most experiments). Unitary evoked EPSPs were monitored continuously at approximately 10 s intervals until peak EPSP amplitudes reached steady-state levels (always within 15 min and usually within approximately 5 min). During data collection, stimulus trains were separated by 6 s to ensure that the system returned to baseline conditions. The interval between stimulus trains was held constant throughout each period of data collection. After data collection, whole cell recording was re-established with pipettes containing 0.5% biocytin for >20 min; slices were fixed immediately and the filled neurons were later visualized using avidin-biotin and 3,3'-diaminobenzidine / horseradish peroxidase histochemistry (DAB/HRP Horikawa and Armstrong 1988), for confirmation of cell type and location.
Drug application
The following compounds were used: D-2-amino-5-phosphonovaleric acid (D-AP5), atropine sulfate, carbamylcholine chloride (carbachol), 6,7-dinitroquinoxaline-2,3(1H,4H)-dione (DNQX), mecamylamine hydrochloride, (-)-nicotine hydrogen tartrate, N-methyl-D-aspartate (NMDA), tetrodotoxin with citrate buffer (TTX) (all purchased from Sigma-Aldrich), AF-DX 116, PD 102807, and pirenzepine dihydrochloride (all purchased from Tocris). For bath application, drugs were prepared as concentrated stock solutions in 0.1 M NaOH (DNQX), 0.1 M HCl (atropine), DMSO (AF-DX 116, PD 102807), or water (all others) and diluted in oxygenated ACSF immediately before use. The final concentration of DMSO was <0.1% in solutions containing AF-DX 116 and PD 102807. Bath-applied antagonists were superfused for 20 min (atropine) or for 10 min (all others), before starting to collect data. For reversible antagonists (D-AP5, mecamylamine, pirenzepine), the order of trials (ACh agonist alone versus ACh agonist plus antagonist) was reversed in successive experiments to negate possible confounding effects of synaptic run-down or receptor desensitization.
For NMDA pressure ejection, data were collected in standard ACSF (1 mM Mg+2/2 mM Ca+2) containing 0.5 µM tetrodotoxin (TTX) and 10 µM DNQX. NMDA-containing pipettes (1–5 MOhm resistance) were positioned 5–20 µm from the apical dendrite, within 50 µm of the soma. NMDA was dissolved in ACSF (pH 8). Pulses of 2–4 ms duration at 20 PSI were delivered using a picospritzer (General Valve, Fairfield, NJ) at 10 s intervals. Current responses were recorded in whole cell configuration as described above, with the soma held in voltage-clamp mode at –60 mV. Responses were digitized at 1 kHz. Average voltage responses were compiled from >20 sweeps.
Data analysis
Individual and average (50–100 sweeps) EPSP traces were analyzed off-line using macros written with Igor software. Experiments where the resting membrane potential (Vm) of either a pre- or postsynaptic cell exceeded –50 mV (where persistent spontaneous firing was observed), or where the access resistance increased substantially (>30%) during the course of the experiment, were excluded from the analysis. Over a given set of trials, resting membrane potential generally did not fluctuate by >2 mV. Injection of holding current in some experiments to hold Vm fixed did not affect the data appreciably. Determinations of peak EPSP amplitude, coefficient of variation (CV) and background noise followed Feldmeyer et al. (1999), except as noted. Mean baseline noise was 0.07 ± 0.02 mV in 1 mM Mg+2 and 0.13 ± 0.05 mV in 0 mM Mg+2. Connections where the first EPSP in an evoked train had an average amplitude of <0.2 mV were excluded due to poor signal:noise ratio (except for data given under "general properties" in the Results, where only connections <0.1 mV were excluded). EPSP traces that contained spontaneous action potentials or large nonevoked EPSPs (>10x mean evoked EPSP amplitude) were excluded. These typically constituted <1% of the data set. For average EPSP traces, individual traces were aligned by triggering to stimulus onset. Triggering to action potential onset (in a subset of the data) yielded results that did not differ appreciably from stimulus-triggered averages. For drug application, a "before" (control) average EPSP was taken from 100 trials immediately preceding drug application; a "during" average was made over the period of peak drug effect (if the effect was unambiguous, e.g., Figure 1A) or else from 50 trials beginning 3 min after drug application (if the effect was absent or ambiguous); an "after" average was taken from 50 trials at the end of the washout period. Mean values are presented ± SD. P values for statistical significance were obtained from 2-tailed, paired t-test except as noted.
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Two hundred forty-five local synaptic connections were found in 3131 unidirectional tests (7.8% connected). Evoked unitary EPSPs were abolished by bath application of 10 µM DNQX (AMPAR antagonist) in conjunction with 50 µM D-AP5 (NMDAR antagonist), indicating that the connections were glutamatergic. Very weak connections (<0.1 mV mean peak amplitude) were not subjected to further data collection. In the remainder of connected pairs, evoked EPSPs had a mean peak amplitude of 0.89 ± 0.74 mV. Mean latency was 1.6 ± 1.1 ms, measured from action potential peak to EPSP onset time. Approximately 75% of connected pairs displayed short-term synaptic depression in response to 10 Hz presynaptic stimulation (cf. Thomson et al. 1993). On recovery and light-microscopic examination of representative fixed biocytin-filled pairs, cells were found to be of the thick tufted pyramidal type (Markram et al. 1997) with apical dendritic branches extending into layer 1 (n = 11 intact pairs and 28 total cells).
Reduction of unitary EPSPs by carbachol
To examine possible modulatory actions of ACh, we first applied the nonhydrolyzable ACh analogue carbachol, which is an agonist for both nicotinic and muscarinic receptors (Brown and Taylor 1996) but was found to produce mainly muscarinic effects in many in vitro reports (e.g., Auerbach and Segal 1996; Fernandez de Sevilla et al. 2002). 5–10 µM carbachol, bath-applied for 2–3 min, did not cause significant or reproducible changes in the resting membrane potential or mean input resistance (Table 1). This application time and concentration range were used throughout the present study. Longer application times and higher concentrations (20–50 µM) were avoided because they caused substantial depolarization and spontaneous firing (cf. McCormick 1992).
|
Reduction of unitary EPSPs by nicotine
To examine nicotinic modulation of EPSPs, 10 µM nicotine was bath-applied for 3 min. The results for a representative pair are shown in Fig. 1C. Compared with the typical effect of carbachol (Fig. 1A), nicotine had relatively little effect on peak EPSP amplitude (also see Table 1). The width of the EPSP (measured at one-half-peak amplitude) did not change significantly (Table 1).
At resting membrane potential, NMDA receptors are largely blocked by Mg+2. This might mask a potential NMDAR-dependent effect of nicotine on synaptic transmission, particularly because nicotine has been reported to selectively enhance transmitter release at NMDAR-only "silent" synapses in developing cortex (Aramakis and Metherate 1998). Conversely, muscarinic receptors have been reported to mediate reduction of transmitter release preferentially at glutamatergic synapses enriched in nonNMDAR in rat hippocampal CA1 pyramidal neurons (Fernandez de Sevilla et al. 2002). To relieve the voltage-dependent block of NMDAR, we substituted an equimolar concentration of Ca+2 (increased from 2 mM to 3 mM) for Mg+2 to maintain a constant divalent cation concentration. This nominally Mg+2-free solution (0 mM Mg+2 ACSF) increased average peak amplitudes by 48% ± 66%, although the change was not statistically significant (P > 0.05, n = 14, not shown) and may have been influenced by increased Ca+2 as well as reduced Mg+2 (cf. Figure 3D). EPSP width increased significantly from 23.5 ± 4.4 ms to 37.9 ± 9.6 ms (P < 0.0005, n = 12). The changes were largely reversed by 50 µM D-AP5, a selective NMDAR antagonist (in 0 mM Mg+2, EPSP amplitude with D-AP5 was 59% ± 5% of control value, P < 0.005, n = 6, not shown).
|
A 10-fold lower concentration (1 µM nicotine; 3–5 min in 0 mM Mg+2 ACSF) had less effect: EPSP amplitude was reduced to 87% ± 28% of control values (n = 8, P > 0.5, not shown). Therefore 10 µM nicotine was used in subsequent experiments.
Pharmacological characterization of carbachol- and nicotine-dependent EPSP reduction
EPSP reduction by carbachol was blocked by 1 µM atropine (Fig. 2A; compare upper and lower panels), indicating the involvement of muscarinic receptors. Similar results were obtained previously with EPSPs evoked with extracellular stimulation of afferents (Hasselmo and Bower 1992; Vidal and Changeux 1993; Gil et al. 1997; Hsieh et al. 2000). Both pirenzepine (an M1 subtype-selective antagonist) and AF-DX 116 (an M2 subtype-selective antagonist) prevented carbachol-dependent EPSP reduction substantially at concentrations >1 µM (Fig. 2B), which are relatively nonselective (cf. Dörje et al. 1991; Daeffler et al. 1999). At lower concentrations, AF-DX 116 appeared to be slightly more effective than pirenzepine (e.g., at 0.1 µM and 1 µM in Fig. 2B). The M4-selective antagonist PD 102807 was completely ineffective at 2 µM, the highest concentration tested (n = 5 pairs, not shown; cf. Augelli-Safran et al. 1998). In contrast to the results with nicotine, removal of Mg+2 from the external medium did not result in any additional suppressive effect of carbachol (EPSP amplitude with carbachol was 78% ± 20% of control EPSP amplitude in 0 mM Mg+2 ACSF, P > 0.05 n = 6, not shown). EPSP reduction by carbachol in 0 mM Mg+2 ACSF was largely prevented by 1 µM atropine (EPSP = 94% ± 22% of control, P > 0.05, n = 9, not shown).
|
|
) rather than, or in addition to, Mg+2-dependent NMDAR unblocking. Prevention of the effect of nicotine by NMDAR antagonist (Fig. 3C) argued against this possibility, but we nevertheless tested it by increasing [Ca+2] while keeping [Mg+2] fixed at 1 mM. In Fig. 3D (top) raising [Ca+2] from 2 mM to 3 mM caused a significant change in EPSP amplitude and in the paired-pulse ratio consistent with an increase in release probability (Johnston and Wu 1995). In ACSF containing 1 mM Mg+2 and 3 mM Ca+2, nicotine had relatively little effect on EPSP amplitude (Fig. 3D, bottom) in comparison to its effect in ACSF containing 0 mM Mg+2 and 3 mM Ca+2 (Fig. 3A). This indicated that the effect of nicotine depended largely on the reduction in [Mg+2] rather than on the increase in [Ca+2]. Because of the nicotinic EPSP reduction appeared to depend on NMDA receptor unblocking, we predicted that the relative magnitude of the effect might increase when NMDAR responses were isolated by bath application of the AMPAR blocker, 10 µM DNQX. Somewhat surprisingly, the effect of nicotine was slightly less pronounced in the presence of DNQX (80% ± 20% of control; P < 0.05, n = 8, not shown).
Effects of carbachol and nicotine on the short-term plasticity of EPSPs
We tested whether the effects of carbachol and nicotine occur through a presynaptic mechanism by examining EPSPs evoked by repetitive presynaptic stimulation. Both carbachol (in 1 mM Mg+2 ACSF) and nicotine (in 0 mM Mg+2 ACSF) increased the paired-pulse ratio, by producing a greater suppressive effect on the first EPSP of a 10 Hz train than on subsequent EPSPs (Fig. 4, A-D; Table 1, "EPSP2/EPSP1"). The change in PPR was statistically significant for both carbachol and nicotine, and was reversible on washout (Fig. 4, C and D). Additional evidence for a presynaptic change was provided by analysis of the change in coefficient of variation relative to mean EPSP amplitude (Fig. 4, E and F; postsynaptic change corresponds to points falling within the shaded regions (Faber and Korn 1991; for applicability of this method in the context of paired intracellular recordings, see Sjöström et al. 2003).
|
). Transient bath application of nicotine had no apparent effect on response amplitude (Fig. 5).
|
The fact that nicotine did not alter postsynaptic response to applied NMDA did not rule out the possibility that postsynaptic NMDAR activity was required for nicotine's effect in a way that was not grossly detectable as a change in the NMDAR current (see DISCUSSION and Fig. 7). Because the Mg+2 block is voltage-dependent, depolarizing the postsynaptic cell (without reducing [Mg+2]) was expected to have an effect similar to removal of Mg+2 if nicotinic EPSP reduction was indeed dependent on the activity of postsynaptic NMDAR. To test this, we held the postsynaptic cell at –20 mV and recording NMDAR-mediated excitatory postsynaptic currents (EPSCs) evoked by presynaptic stimulation (Fig. 6). A fraction of synaptic contacts, at least, between connected pairs was electrotonically close enough to the soma to permit unblocking of NMDAR by this method (compare Markram et al. 1997). Bath-application of nicotine caused changes in EPSC amplitude (Fig. 6, A-C) and paired-pulse ratio (Fig. 6C) that were comparable to the effects of nicotine in 0 mM Mg+2 ACSF (compare Fig. 6 to Figs. 1, E and F, 3A, 4, B and D).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Receptor types and molecular mechanisms underlying EPSP reduction
Muscarinic receptors.
Reduction of EPSPs by carbachol in 1 mM Mg+2 was largely blocked by 1 µM atropine (Fig. 2A), indicating that the effect depends on muscarinic receptors. Muscarinic EPSP reduction has been seen in studies of many CNS pathways, but the receptor subtype(s) responsible have not been unambiguously identified. Vidal and Changeux (1993) found that muscarinic reduction of EPSPs evoked by extracellular stimulation in prefrontal cortex was blocked less effectively by the M1-selective antagonist pirenzepine than by the nonselective antagonist atropine, suggesting that receptors other than the M1 subtype were responsible. In immunocytochemical studies encompassing rat (Levey et al. 1991), primate (Mrzljak et al. 1993; Mrzljak et al. 1996; Mrzljak et al. 1998) and cat sensory cortex (Erisir et al. 2001), M2 receptor staining was consistent with both pre- and postsynaptic distribution at the light microscopic level; electron microscopy has identified M2 in axon terminals and dendritic spines forming asymmetric (presumed glutamatergic) synapses (Mrzljak et al. 1993; Mrzljak et al. 1996; Erisir et al. 2001). In contrast, M1 receptors are located postsynaptically in cortical pyramidal cells in rat (Levey et al. 1991) and primate cortex (Mrzljak et al. 1993), but have not been seen presynaptically. In support of a role for M2 receptors, our data suggest that the M2 antagonist AF-DX 116 prevents EPSP reduction by carbachol somewhat more effectively than the M1 antagonist, pirenzepine (Fig. 2B). Other than M1 and M2, M4 is the most abundant muscarinic receptor subtype in rat brain (Levey et al. 1991). But high concentrations of the M4-specific antagonist PD 102807 did not affect carbachol-mediated EPSP suppression, so involvement of M4 subunits appears unlikely.
In 0 mM Mg+2, carbachol might have been expected to cause additional EPSP reduction via nicotinic receptors, based on the fact that nicotine reduced EPSPs in 0 mM Mg+2 but not in 1 mM Mg+2 (Fig. 1, C-F). However, the effect of carbachol in 0 mM Mg+2 was not significantly different from its effect in 1 mM Mg+2, and was blocked by atropine. This indicates that carbachol acted as a de facto muscarinic agonist in our system, consistent with several previous reports (see RESULTS).
Nicotinic receptors
Previous electrophysiological studies of nicotinic modulation of cortical glutamatergic synapses have found either no effect, or a strictly facilitative effect (Vidal and Changeux 1993; Aramakis and Metherate 1998) (Gil et al. 1997). Our finding that nicotine can reversibly reduce glutamatergic EPSPs therefore reveals a previously unrecognized activity of nicotine in sensory cortex; (but see Maggi et al. 2004; discussed in the following text). The difficulty of selectively activating intra-cortical pathways by extracellular stimulation may explain why the suppressive action of nicotine was not observed previously. Moreover, our results show that conditions favoring NMDAR activation are necessary to detect nicotine-dependent EPSP reduction in slice recordings. Another possibility is that EPSP reduction by nicotine might be a feature of local but not longer-distance intra-cortical connections. Analogously, nicotine might enhance EPSPs in the superficial layers, such as connections between layer 4 spiny stellate cells and layer 2/3 pyramidal cells, while reducing EPSPs between layer 5 pyramidal cells. In this scheme, nicotinic activity would differently modulate early versus late stages of processing within a cortical column. Also, the effect might be developmentally transient, because our data were taken from animals within or near the critical period for sensory plasticity (Fox 1992; Glazewski and Fox 1996).
We conclude that nicotinic EPSP reduction happens through presynaptic reduction of transmitter release, based on the analysis of paired-pulse ratio (Fig. 4, A-D) and coefficient of variation (Fig. 4, E and F). Nicotinic EPSP reduction also required NMDAR unblocking via removal of Mg+2 from the external medium (Fig. 1, C-F), or depolarization of the postsynaptic cell (Fig. 6). In addition, nicotinic EPSP reduction was prevented by the NMDAR antagonist, D-AP5 (Fig. 3, A and B). One possible explanation for these observations is that nicotine preferentially reduces glutamate release at functionally "silent" sites that lack detectable AMPAR. Nicotinic receptors might be located selectively at terminals presynaptic to NMDAR-only sites (Fig. 7A). An analogous mechanism has been proposed for NMDAR-selective nicotinic enhancement of EPSPs evoked in presumed thalamo-cortical synapses of developing auditory cortex (Aramakis and Metherate 1998), and for apparent AMPAR-selective muscarinic modulation in hippocampus (Fernandez de Sevilla et al. 2002).
However, not all our data fit this model. If nicotinic modulation in our system targets postsynaptic sites enriched in NMDAR, EPSP reduction (normalized to control) would be more prominent when nonNMDAR are blocked by DNQX. Instead, we found that nicotinic reduction of the isolated NMDAR response, either in 0 mM Mg+2 (not shown) or with the postsynaptic cell depolarized under voltage clamp (Fig. 6) was not conspicuously greater than nicotinic reduction of the combined NMDAR+AMPAR response (Fig. 1, E and F, 3A). Quantitative comparisons across different conditions may be of limited validity in our system because the effects we measured were transient, reflecting limited bath application time and possible receptor desensitization, particularly of nicotinic receptors (reviewed in Giniatullin et al. 2005). But taken at face value, these data argue against selective reduction of the AMPAR-dependent component.
Therefore an alternative explanation for our findings is that glutamate release is not modulated selectively at postsynaptically NMDAR-enriched sites, but instead that modulation of release occurs at mixed AMPAR+NMDAR sites, and is a result of a change in NMDAR activity (Fig. 7B). The details of this model are more speculative at present than the details of the first model (Fig. 7A). Our results suggest an obligatory role for postsynaptic NMDAR, because depolarization of the postsynaptic neuron elicits nicotine-dependent EPSP reduction (Fig. 6). We found no grossly appreciable effect of nicotine on currents produced by exogenous NMDA application (Fig. 5), although it is possible that a selective synaptic effect was masked by responses of extrasynaptic NMDAR, whose subunit composition may differ from that of synaptic NMDAR (Stocca and Vicini 1998; Tovar and Westbrook 1999), and which might not respond to nicotinic modulation in the same way as synaptic NMDAR. There is evidence that nicotinic activity can affect postsynaptic NMDAR activity by a calmodulin-dependent process (Fisher and Dani 2000). Postsynaptic NMDAR modulation could then affect the presynaptic terminal via a diffusible agent such as nitric oxide; nitric oxide synthase has been identified in cortex within dendritic spines that contain NMDAR (Aoki et al. 1997) and nitric oxide has been linked to NMDAR activation and modulation of presynaptic glutamate release (Montague et al. 1994; Garthwaite and Boulton 1995). The model does not exclude a role for presynaptic NMDAR, which have been identified by electron microscopic immunocytochemistry in rat visual cortex (Aoki et al. 1994) and elsewhere in cortex (DeBiasi et al. 1996; Charton et al. 1999), and have been implicated in timing-based long-term depression at synapses between layer 5 pyramidal cells of rat visual cortex (Sjöström et al. 2003; cf. Berretta and Jones 1996).
It is also possible that nicotine reduces EPSPs indirectly by enhancing GABA release from inhibitory neurons (see references cited in the INTRODUCTION), with consequent reduction of glutamate release (Bonanno et al. 1997; Torres-Escalante et al. 2004). We cannot rule out this possibility, although some indirect evidence argues against it. Baseline spontaneous activity in our recordings was generally low, and we did not find evidence for spontaneous inhibition emerging as a result of agonist application, although inhibitory PSPs may have been difficult to detect due to the composition of our solutions. Additionally, we have recorded from both fast-spiking (FS) and low-threshold-spiking (LTS) interneurons, alone and also receiving synaptic input with pyramidal neurons (n = 10 pairs for FS and 4 pairs for LTS), and carbachol or nicotine application under the conditions of the present study did not elicit spontaneous or evoked firing (via stimulation of a presynaptic pyramidal neuron) in either interneuron type.
A recent study of using slices from neonatal (P1-P7) rats found that transient nicotine application produced long-term depression (LTD) of EPSCs evoked in hippocampal CA1 pyramidal cells by stimulation of Schaffer collaterals (Maggi et al. 2004). Unlike the results of that study, nicotinic EPSP reduction in the present case was largely reversible. Maggi et al. (2004) also found that nicotine can either potentiate or depress synaptic transmission, depending on whether transmitter release probability (P) was low or high, respectively; changing P by increasing extracellular [Ca+2] and reducing [Mg+2] could convert the effect of nicotine at a given synapse from LTP induction to LTD induction, or vice versa. In the present study, increasing [Ca+2] without reducing [Mg+2] did not elicit nicotine-dependent EPSP reduction (Fig. 3D). We could not determine P directly in most instances because it was difficult to distinguish very weak EPSPs from failures, but all the connections in our data set were unambiguously reliable (P > 0.5), It is possible that low P connections in respond to nicotine differently from those we have studied here.
NMDAR activity is reduced but not absent in 1 mM Mg+2, relative to 0 mM Mg+2; the NMDAR component has been reported to compose 17% of the voltage-time interval of the EPSP at –60 mV in connections between layer 5 pyramidal neurons in slices from P14-P16 rats, recorded in 1 mM Mg+2 (Markram et al. 1997). Based on this, we would expect (in principle) some degree of EPSP reduction by nicotine even in 1 mM Mg+2, whereas our pooled data showed no such effect (Fig. 1D). We speculate that the effect of nicotine under these conditions was too small to be statistically evident, given the limited number of samples (n = 41 pairs). A suggestion of nicotinic EPSP suppression in 1 mM Mg+2 ACSF was seen in some individual instances (e.g., Fig. 1C).
Sources of variability in the data
We did not find correlations between modulatory effects and other parameters such as age of the animal, baseline EPSP amplitude or PPR, so the foregoing discussion all pairs were assumed to belong to a homogeneous population. However, the data give some indication of variation in modulatory effect among pairs. For instance, the histogram of nicotine's effect in 1 mM Mg+2 is approximately Gaussian and centered on 1 (Fig. 1D, right), but some individual pairs showed a reversible EPSP enhancement by nicotine while others showed a reversible reduction (Fig. 1D, left). In view of this, it remains possible that EPSP modulation is correlated with differences in cell type or connectivity that were not examined here. For example, we did not make a morphometric distinction between neurons with simple versus complex apical dendritic tufts, both of which classes may be regular-spiking (Franceschetti et al. 1998) but may differ in their synaptic properties (Angulo et al. 2003). Individual pairs within layer 5 could also represent different circuit elements, e.g., intra- versus intercolumnar connections could be modulated differently. Possible functional consequences are discussed in the next section.
Cholinergic reduction of local excitatory transmission: functional aspects
It has been proposed that recurrent excitation in the visual cortex influences neuronal selectivity and responsiveness to stimulus properties including orientation (Ben-Yishai et al. 1995; Douglas et al. 1995; Somers et al. 1995), direction (Douglas et al. 1995) and spatial phase (Chance et al. 1999). Suppression of local excitatory connections by ACh or other cholinergic agonists could alter the receptive field properties of cortical sensory neurons. Depending on the level of cholinergic activity, the muscarinic pathway would act to reduce the effects of recurrent excitation: lateral spread of excitation, and cortical amplification of sensory input.
The nicotinic pathway would act similarly, but with an additional feature: because of its dependence on NMDA receptor activation, nicotinic reduction of recurrent excitation would be turned on by heightened sensory excitation. Such activity-dependent nicotinic suppression could serve as a cellular mechanism underlying the contraction of receptive field size in macaque V1 in response to increased stimulus contrast (Sceniak et al. 1999). The latter effect was not correlated with changes in surround inhibition, leading the authors to propose a contrast-dependent suppression of lateral excitatory connections. The changes of cortical receptive field properties caused by either the muscarinic or the nicotinic pathway could contribute to the improvements in perceptual ability that have been attributed to ACh (reviewed in Hasselmo 1995). However, our slice data were derived from relatively young animals. It remains to be seen whether nicotinic EPSP reduction persists in adulthood. If it is restricted to the first few weeks after birth, nicotinic EPSP reduction could still be significant in vivo, by shaping cortical receptive field properties during the critical period.
EPSP reduction by both nicotine and carbachol in our system acted by a presynaptic mechanism and thus had a decreasing effect on the amplitude of each successive EPSP in a train. As a result, this type of synaptic modulation might have subtle effects on information coding (Abbott et al. 1997; Markram and Tsodyks 1996a, b; Tsodyks and Markram 1997), rather than simply decreasing communication among locally connected neurons.
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: R. B. Levy, New York University Center for Neural Science, 4 Washington Place, Room 809, New York, NY 10003 (E-mail: rlevy{at}cns.nyu.edu)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Angulo MC, Staiger JF, Rossier J, and Audinat E. Distinct local circuits between neocortical pyramidal cells and fast-spiking interneurons in young adult rats. J Neurophysiol 89: 943–953, 2003.
Aoki C, Rhee J, Lubin M, and Dawson TM. NMDA-R1 subunit of the cerebral cortex co-localizes with neuronal nitric oxide synthase at pre- and postsynaptic sites and in spines. Brain Res 750: 25–40, 1997.[CrossRef][ISI][Medline]
Aoki C, Venkatesan C, Go CG, Mong JA, and Dawson TM. Cellular and subcellular localization of NMDA-R1 subunit immunoreactivity in the visual cortex of adult and neonatal rats. J Neurosci 14: 5020–5222, 1994.
Aramakis VB and Metherate R. Nicotine selectively enhances NMDA receptor-mediated synaptic transmission during postnatal development in sensory neocortex. J Neurosci 18: 8485–8495, 1998.
Ashe JH, McKenna TM, and Weinberger NM. Cholinergic modulation of frequency receptive fields in auditory cortex II. Frequency-specific effects of anticholinesterases provide evidence for a modulatory action of endogenous ACh. Synapse 4: 44–54, 1989.[CrossRef][ISI][Medline]
Auerbach JM and Segal M. Muscarinic receptors mediating depression and long-term potentiation in rat hippocampus. J Physiol 492: 479–493, 1996.
Augelli-Szafran CE, Jaen JC, Moreland DW, Nelson CB, Penvose-Yi JR, and Schwarz RD. Identification and characterization of m4 selective muscarinic antagonists. Bioorg Med Chem Lett 8: 1991–1996, 1998.[CrossRef][Medline]
Ben-Yishai R, Bar-Or RL, and Sompolinsky H. Theory of orientation tuning in visual cortex. Proc Natl Acad Sci USA 92: 3844–3848, 1995.
Berretta N and Jones RSG. Tonic facilitation of glutamate release by presynaptic N-methyl-D-aspartate autoreceptors in the entorhinal cortex. Neuroscience 75: 339–344, 1996.[CrossRef][ISI][Medline]
Bonanno G, Fassio A, Schmid G, Severi P, Sala R, and Raiteri M. Pharmacologically distinct GABAB receptors that mediate inhibition of GABA and glutamate release in human neocortex. Br J Pharmacol 120: 60–64, 1997.[ISI]
Brown JH and Taylor P. Muscarinic receptor agonists and antagonists. In: The Pharmacological Basis of Therapeutics, 9th Edition, edited by Hardman JG, Gilman AG, and Limbird LA. New York: McGraw-Hill, 1996. p. 146–160.
Chance FS, Abbott LF, and Reyes AD. Gain modulation from background synaptic input. Neuron 35: 773–782, 2002.[CrossRef][ISI][Medline]
Chance FS, Nelson SB, and Abbott LF. Complex cells as cortically amplified simple cells. Nat Neurosci 2: 277–282, 1999.[CrossRef][ISI][Medline]
Charton JP, Herkert M, Becker CM, and Schröder H. Cellular and subcellular localization of the 2B-subunit of the NMDA receptor in the adult rat telencephalon. Brain Res 816: 609–617, 1999.[CrossRef][ISI][Medline]
Christophe E, Roebuck A, Staiger JF, Lavery DJ, Charpak S, and Audinat E. Two types of nicotinic receptors mediate an excitation of neocortical layer I interneurons. J Neurophysiol 88: 1318–1327, 2002.
Daeffler L, Schmidlin F, Gies J-P, and Landry Y. Inverse agonist activity of pirenzepine at M2 muscarinic acetylcholine receptors. Br J Pharmacol 126: 1246–1252, 1999.[CrossRef][ISI]
DeBiasi S, Minelli A, Melone M, and Conti F. Presynaptic NMDA receptors in the neocortex are both auto- and heteroreceptors. Neuroreport 7: 2773:2776, 1996.[ISI][Medline]
Descarries L, Krnjevic K, and Steriade M, eds. Acetylcholine in the Cerebral Cortex. Amsterdam: Elsevier, 2004.
Deuchars J, West DC, and Thomson AM. Relationships between morphology and physiology of pyramid-pyramid single axon connections in rat neocortex in vitro. J Physiol 478 Pt 3: 423–435, 1994.
Dörje F, Wess J, Lambrecht G, Tacke R, Mutschler E, and Brann MR. Antagonist binding profiles of five cloned human muscarinic receptor subtypes. J Pharmacol Exp Ther 256: 727–733, 1991.
Douglas RJ, Koch C, Mahowald M, Martin KA, and Suarez HH. Recurrent excitation in neocortical circuits. Science 269: 981–985, 1995.
Erisir A, Levey AI, and Aoki C. Muscarinic receptor M(2) in cat visual cortex: laminar distribution, relationship to gamma-aminobutyric acidergic neurons, and effect of cingulate lesions. J Comp Neurol 441: 168–185, 2001.[CrossRef][ISI][Medline]
Faber DS and Korn H. Applicability of the coefficient of variation method for analyzing synaptic plasticity. Biophys J 60: 1288–1294, 1991.
Feldmeyer D, Egger V, Lubke J, and Sakmann B. Reliable synaptic connections between pairs of excitatory layer 4 neurones within a single ‘barrel’ of developing rat somatosensory cortex. J Physiol 521 Pt 1: 169–190, 1999.
Fernandez de Sevilla D, Cabezas C, Oshima de Prada A, Sanchez-Jimenez A, and Buno W. Selective muscarinic regulation of functional glutamatergic Schaffer collateral synapses in rat CA1 pyramidal neurons. J Physiol 545: 51–63, 2002.
Fisher JL and Dani JA. Nicotinic receptors on hippocampal cultures can increase synaptic glutamate currents while decreasing the NMDA-receptor component. Neuropharmacol 39: 2756–2769, 2000.[CrossRef][ISI][Medline]
Fox K. A critical period for experience-dependent synaptic plasticity in rat barrel cortex. J Neurosci 12: 1826–1838, 1992.[Abstract]
Franceschetti S, Sancini G, Panzica F, Radici C, and Avanzini G. Postnatal differentiation of firing properties and morphological characteristics in layer V pyramidal neurons of the sensorimotor cortex. Neuroscience 83: 1013–1024, 1998.[CrossRef][ISI][Medline]
Garthwaite J and Boulton CL. Nitric oxide signaling in the central nervous system. Ann Rev Physiol 57: 683–706, 1995.[CrossRef][ISI][Medline]
Gil Z, Connors BW, and Amitai Y. Differential regulation of neocortical synapses by neuromodulators and activity. Neuron 19: 679–686, 1997.[CrossRef][ISI][Medline]
Giniatullin R, Listri A, and Yakel JL. Desensitization of nicotinic ACh receptors: shaping cholinergic signaling. Trends Neurosci 28: 371–378, 2005.[CrossRef][ISI][Medline]
Glazewski S and Fox K. Time course of experience-dependent synaptic potentiation and depression in barrel cortex of adolescent rats. J Neurophysiol 75: 1714–1729, 1996.
Hasselmo ME. Neuromodulation and cortical function: modeling the physiological basis of behavior. Behav Brain Res 67: 1–27, 1995.[CrossRef][ISI][Medline]
Hasselmo ME and Bower JM. Cholinergic suppression specific to intrinsic not afferent fiber synapses in rat piriform (olfactory) cortex. J Neurophysiol 67: 1222–1229, 1992.
Horikawa K and Armstrong WE. A versatile means of intracellular labeling: injection of biocytin and its detection with avidin conjugates. J Neurosci Methods 25: 1–11, 1988.[CrossRef][ISI][Medline]
Hsieh CY, Cruikshank SJ, and Metherate R. Differential modulation of auditory thalamocortical and intracortical synaptic transmission by cholinergic agonist. J Neurosci 20: 6106–6116, 2000.
Johnston D and Wu SM. Foundations of Cellular Neurophysiology. Cambridge, MA: MIT Press, 1995.
Kawaguchi Y. Selective cholinergic modulation of cortical GABAergic cell subtypes. J Neurophysiol 78: 1743–1747, 1997.
Kimura F, Fukuda M, and Tsumoto T. Acetylcholine suppresses the spread of excitation in the visual cortex revealed by optical recording: possible differential effect depending on the source of input. Euro J Neurosci 11: 3587–3609, 1999.
Krnjevic K, Pumain R, and Renaud L. The mechanism of excitation by acetylcholine in the cerebral cortex. J Physiol 215: 247–268, 1971.
Levey AI, Kitt CA, Simonds WF, Price DL, and Brann MR. Identification and localization of muscarinic acetylcholine receptor proteins in brain with subtype-specific antibodies. J Neurosci 11: 3218–
6 min) during the period of greatest EPSP suppression following the start of carbachol application (5 µM). Resting Vm = –66 mV; baselines are aligned for comparison. Right, time course of the carbachol-mediated effect on the amplitude of unitary EPSPs. Carbachol was applied for 3 min (bar). Open circles denote the peak amplitudes of unitary EPSPs; the solid line is a sliding boxcar average of the same data (width = 19 data points). B: data for all pairs. Left, average amplitudes before (control) and during agonist application, and after washout. Data are normalized to control values. Gray lines and open circles show values for individual pairs. Black lines and filled circles with error bars show mean and SD; values are 0.71 ± 0.24 (carbachol) and 0.98 ± 0.23 (washout). The asterisk denotes significance of average agonist value relative to the control (P < 0.0001, n = 39) and washout values (P < 0.0001, n = 36; washout data were not obtained for three pairs). Right, histograms showing EPSP amplitude reduction by carbachol. The open circle with bars shows the mean ± SD. C and D: effect of nicotine on evoked unitary EPSP amplitude in 1 mM Mg+2 ACSF. C: left, evoked EPSPs before (control) and during bath application of nicotine (10 µM, 3 min). Resting Vm = –62 mV. Right, time course of EPSP amplitude (open circles) and sliding average. D: data for all pairs. The normalized EPSP amplitude during nicotine is 1.02 ± 0.22 and the washout value is 1.01 ± 0.29 (n = 41). E: effect of 10 µM nicotine on the same pair as in B, in 0 mM Mg+2 ACSF. Resting Vm = –61 mV. (Note: the broadening of the EPSP during nicotine in this trace was not typical; compare traces in 
