INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In several brain regions, the strength of glutamate-dependent neurotransmission can be enhanced via nicotinic acetylcholine receptors (nAChRs), primarily through the highly Ca²⁺-permeable 7 subtype (Albuquerque et al. 1997; Alkondon et al. 1996; Aramakis and Metherate 1998; Gil et al. 1997; Gray et al. 1996; Ji et al. 2001; McGehee et al. 1995); however, in certain brain areas, non-7 nAChRs affect glutamate transmission (Gil et al. 1997; Guo et al. 1998; Vidal and Changeux 1993). Nicotine-induced long-term potentiation has been attributed to its action on 7 nAChRs present on glutamate terminals (Fujii et al. 2000; Mansvelder and McGehee 2000). In the rat hippocampus, pyramidal neuron excitatory postsynaptic current (EPSC) is enhanced by 7 nAChR activation (Gray et al. 1996; but see Vogt and Regehr 2001); however, it is not known whether interneuron EPSCs are regulated by 7 or any other nAChRs. Because glutamate EPSCs remain the major excitatory stimulus to the interneurons (McBain et al. 1999), we tested the effects of activation of nAChRs on interneuron EPSCs in the hippocampus and attempted to determine pharmacologically the nature of the nAChR subtype involved.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Whole-cell patch-clamp experiments were performed on CA1 stratum radiatum (SR) interneurons of rat hippocampal slices obtained from the brain of 15- to 24-day-old Sprague-Dawley rats. All recording conditions, solutions, and analyses were as previously described (Alkondon et al. 2000). The slices were superfused with artificial cerebrospinal fluid (ACSF) at 2 ml/min in the presence of atropine (1 µM) to block the muscarinic receptors. All recordings were performed at room temperature (20-22°C). Agonists were applied via a U-tube, and antagonists were applied via bath superfusion (Alkondon et al. 2000). Data are expressed as the mean ± SE.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In the absence of any antagonists, both EPSCs and inhibitory postsynaptic currents (IPSCs) were recorded from CA1 SR interneurons at 68 mV (Fig. 1A). The EPSCs could be blocked by 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX, 10 µM), whereas the IPSCs were sensitive to bicuculline (5 or 10 µM). Glutamate EPSCs were short in duration (rise time = 1.08 ± 0.06 ms; _decay = 3.8 ± 0.34 ms; n = 9), and GABAergic IPSCs longer (rise time = 2.83 ± 0.20 ms; _decay = 14.6 ± 1.29 ms; n = 9). Following this initial characterization, bicuculline (5 or 10 µM) was routinely included in the ACSF to block IPSCs that allowed the study of EPSCs in isolation.

View larger version (28K): [in this window] [in a new window]   Fig. 1. Acetylcholine (ACh) enhances excitatory postsynaptic current (EPSC) frequency in the CA1 stratum radiatum (SR) interneurons. A: samples of averaged traces of spontaneous EPSCs and inhibitory postsynaptic currents (IPSCs) collected in the absence of glutamate and GABA receptor blockers. Events were separated by visual inspection of postsynaptic currents (PSCs). B: sample current traces from an SR interneuron to which U-tube pulses of artificial cerebrospinal fluid (ACSF) or ACh (100 µM) were applied for 12 s (solid line above the top trace). Inset traces represent averaged EPSCs (calibration = 10 ms). The AMPA receptor antagonist 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX, 10 µM) was applied for 8 min. Bicuculline (10 µM) was present throughout. C: cumulative distribution plots of the amplitude of the EPSCs and the inter-event intervals. The ACh-induced shift in the distribution of both events is statistically significant (P < 0.001) by the Kolmogorov-Smirnov test. All recordings were done at 68 mV.

U-tube application of a short pulse of acetylcholine (ACh; 100 µM) increased the frequency of EPSCs in 72 of 104 CA1 SR interneurons (Fig. 1B). CNQX (10 µM) abolished both spontaneous and ACh-triggered EPSCs (Fig. 1B). Although ACh increased significantly (P < 0.001 by Student's paired t-test) the frequency of EPSCs (n = 72 neurons) from a control value of 0.09 ± 0.011 to 0.76 ± 0.074 Hz, neither the rise time (0.99 ± 0.05 ms in control to 1.07 ± 0.05 ms in ACh) nor the _decay (2.83 ± 0.15 ms in control to 2.70 ± 0.14 ms in ACh) of the EPSCs was changed by the agonist. In 54 of 72 neurons, ACh also significantly enhanced the mean peak amplitude of EPSCs (P < 0.001) from a value of 17.2 ± 0.9 to 25.0 ± 1.5 pA (Fig. 1B). In the other 18 neurons, the ACh-induced increase in the frequency of EPSCs was not accompanied by a change in the peak amplitude of the currents. ACh-induced changes were also evident in the cumulative plots of the inter-event interval and the amplitude of the events (Fig. 1C).

To investigate which nAChR subtype might be involved in the modulation of glutamate transmission to the interneurons, different nicotinic agonists were compared with respect to their effectiveness to increase the frequency of EPSCs. It has been shown that a preferential activation of certain nAChR subtypes occurs with choline (7), dimethyl phenyl piperazinium (DMPP; 32), and cytisine (34) (Albuquerque et al. 1997; Luetje and Patrick 1991; Quick et al. 1999), and therefore, a rather reliable correlation can be made between sensitivity to agonist and nAChR subtype. U-tube application of nicotine (100 µM; 6-12 s) triggered a short burst of EPSCs, including some with large amplitudes that were rarely observed under control (Fig. 2A). In neurons (n = 6) in which ACh (100 µM) elicited both nicotinic inward current and enhanced frequency of EPSCs, choline (1 mM; 12 s) induced a slight increase in the EPSC frequency (1 of 6 cells; Fig. 2B) or primarily the nicotinic current only (Fig. 2C). A higher concentration of choline (10 mM; n = 12) evoked large-amplitude nicotinic currents, but did not increase the frequency of EPSCs. Comparison of the agonists in the same neurons, as shown in Fig. 2D, indicated that cytisine was the most effective agonist for inducing EPSCs. An analysis of the frequency changes in various neurons indicated a statistically significant increase in the number of EPSCs (Fig. 2E) by four agonists with an order of effectiveness as follows: cytisine > nicotine > DMPP > ACh. Bath application of low concentrations of nicotine (250 and 500 nM) also enhanced EPSC frequency (Fig. 2F).

View larger version (31K): [in this window] [in a new window]   Fig. 2. Nicotinic agonists enhance differentially the EPSC frequency. A-D: sample current traces from 4 interneurons showing the effect of different agonists. Inset in A shows averaged EPSCs in the presence of nicotine. E: summary of the EPSC frequency changes induced by various agents. Control frequency was calculated for 3-4 min before applying the agents. Frequency of EPSCs during application of various agents was calculated for 30 s from the onset of the agonist pulse. The number of neurons was 9 (ACSF), 72 (ACh), 9 (nicotine), 6 (choline), 8 each for dimethyl phenyl piperazinium (DMPP) and cytisine. *P < 0.001 by paired t-test, compared with control. F: bar graph shows the average changes in the frequency of EPSCs induced by bath-applied nicotine from 4 neurons (250 nM) and 7 neurons (500 nM). *P < 0.05 by paired t-test.

Nicotinic antagonists are also useful in distinguishing between various nAChR subtypes (Albuquerque et al. 1997). Methyllycaconitine (MLA; 10 nM) and -bungarotoxin (BGT; 50 nM), two potent antagonists of 7 nAChR, failed to inhibit the agonist-induced increase in the frequency of EPSCs (Fig. 3, A, B, and F). Dihydro--erythroidine (DHE; 10 µM), a potent inhibitor of 42 nAChRs, was more effective in inhibiting nicotinic currents than ACh-induced EPSCs (Fig. 3, C and F). However, mecamylamine (MEC; 10 µM), a broad-spectrum nicotinic antagonist that has higher potency for 34 nAChRs than for other nAChR subtypes (Papke et al. 2001), inhibited about 95% of the EPSCs induced by the agonist (Fig. 3, D and F). Further, TTX (0.5 µM), a Na⁺ channel blocker, reduced significantly ACh-mediated increase in the EPSC frequency, but did not affect the nicotinic current (Fig. 3, E and F). In the presence of TTX (n = 6), ACh did not trigger any large-amplitude EPSCs.

View larger version (31K): [in this window] [in a new window]   Fig. 3. Differential inhibition of nicotinic agonist-mediated enhancement of EPSC frequency. A-E: sample recordings of agonist-triggered EPSCs from 5 SR interneurons before exposure and 10 min (methyllycaconitine, MLA, 10 nM; dihydro--erythroidine, DHE, 10 µM; mecamylamine, 10 µM, TTX, 0.5 µM) or 15 min (-bungarotoxin, BGT; 50 nM) after exposure to the blockers. F: summary of the EPSC frequency changes induced by various blockers (n = 5 to 6). *P < 0.001 by one-way analysis of variance (ANOVA) with Tukey's multiple comparison test, compared with agonist group.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The main finding of this study is that a non-7 nAChR, probably 34, which can be activated by low concentrations of nicotine, regulates glutamate synaptic transmission to CA1 interneurons. The strong agonist response to cytisine, a potent inhibitory response to mecamylamine, and a poor inhibitory response to DHE of EPSCs recorded from CA1 SR interneurons, suggest the involvement of 34-containing nAChRs and negate the involvement of 42-containing nAChRs. The present results, however, cannot exclude the participation of other potential combinations of subunits (e.g., 324) that are yet to be confirmed. In addition, we deduce based on a weak choline response and a MLA-insensitive ACh response that an 7 nAChR-dependent EPSC modulation is negligible in the CA1 region of the SR interneurons that contrasts with the previous reports of the presence of an 7 nAChR-dependent EPSC modulation in CA1 pyramidal neurons (Gray et al. 1996; Ji et al. 2001). Thus we propose the existence of a neuron-specific control of excitatory inputs by distinct subtypes of nAChRs in the hippocampus, i.e., an 7 regulating the function of pyramidal neurons and a non-7, presumably 34, regulating interneurons (see Fig. 4). Such a dichotomy can add to the flexibility the brain ought to have to achieve fine tuning of several functions.

View larger version (23K): [in this window] [in a new window]   Fig. 4. Circuit diagram of CA1 hippocampal neurons showing the potential role of different nicotinic acetylcholine receptor (nAChR) subtypes. The 7- and 42-containing nAChRs present on the GABAergic interneurons exert inhibition and disinhibition of pyramidal neurons (see Alkondon and Albuquerque 2001). Presynaptic 7 nAChRs on the glutamate terminals (Alkondon et al. 1996; Gray et al. 1996) mediate excitation of pyramidal neurons. A non-7 nAChR (34?) present on the dendritic region of a pyramidal neuron (Ji et al. 2001) or on the preterminal axonal region (thick line near the axon terminal) of another unidentified glutamate neuron causes excitation of interneuron via glutamate release.

The nAChR subtypes that modulate EPSCs differ in their location and the mechanism by which they affect the transmitter release. For instance, 7 nAChRs in many systems are located at glutamate terminals where they mediate release solely through a Ca²⁺-dependent mechanism (see INTRODUCTION), increasing the frequency but not the amplitude of EPSCs. The ability of TTX to abolish nAChR-triggered EPSCs in the interneurons (see Fig. 3) suggests that the non-7 nAChR is not located on the terminals. Rather, the observation that the peak amplitude of EPSCs is increased by the nicotinic agonists (see Fig. 1) is consistent with a non-7 nAChR-dependent mechanism that involves either an action potential initiation or a Na⁺ channel-dependent depolarization that propagates across several terminal branches, resulting in a synchronized transmitter release at multiple release sites. The observation that ACh failed to increase the amplitude of EPSCs in the presence of TTX further rules out any effect of the nicotinic agonist on the sensitivity of postsynaptic glutamate receptors. The source of non-7 nAChR-dependent glutamate input to SR interneuron can be either a pyramidal neuron, which exhibits a mecamylamine-sensitive nicotinic current (Ji et al. 2001), or a hitherto unidentified neuron or axon present in the hippocampal slices (see Fig. 4). The TTX sensitivity of the responses suggest a possible preterminal location (see Fig. 4) on the glutamatergic axon for the non-7 nAChR.