Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 1999 Jul 1;518(Pt 1):131–140. doi: 10.1111/j.1469-7793.1999.0131r.x

Activation of nicotinic acetylcholine receptors patterns network activity in the rodent hippocampus

Stuart R Cobb 1, Diederik O Bulters 1, Sachin Suchak 1, Gernot Riedel 1, Richard G M Morris 1, Ceri H Davies 1
PMCID: PMC2269408  PMID: 10373695

Abstract

  1. Intracellular and extracellular recordings from area CA3 of rat and mouse hippocampal slices revealed two distinct modes of synchronous network activity in response to continuous application of muscarinic acetylcholine receptor (mAChR) agonists. At low concentrations (e.g. 0·1-1 μM oxotremorine-M), ‘burst-mode’ activity comprised regular individual AMPA receptor-mediated depolarizing events, each generating several action potentials. At higher concentrations (5-50 μM), ‘theta-mode’ prevailed in which ordered clusters of depolarizing theta-frequency oscillations occurred.

  2. Whilst theta-mode activity was abolished by the mAChR antagonist atropine (5 μM), the nicotinic acetylcholine receptor (nAChR) antagonists tubocurarine (100 μM), mecamylamine (100-500 μM) and dihydro-β-erythroidine (250 μM) converted this mode of activity to burst-mode.

  3. Likewise, disruption of synaptically available ACh using inhibitors of choline uptake (hemicholinium-3; 20-50 μM) or vesicular ACh transport (vesamicol; 50 μM) converted theta-mode into burst-mode activity.

  4. Hippocampal slices prepared 2-3 weeks after transection of the primary cholinergic efferent pathway from the medial septum exhibited reduced vesicular ACh transporter immunoreactivity but still supported nAChR-dependent theta-mode activity suggesting that ACh released from this pathway was not critical for the activation of these receptors.

  5. In summary, ACh-mediated activation of nAChRs tailors the pattern of network activity into theta-frequency depolarizing episodes as opposed to synchronized individual events at much lower frequencies.

Synchronization of cortical network activity is believed to underlie a variety of physiological as well as pathophysiological functions (Buzsáki et al. 1983, 1994; Singer, 1993; Jefferys et al. 1996; Paulsen & Moser, 1998). To date, most studies have concentrated on the hippocampus where a variety of patterns of rhythmic activity have been recorded in vitro in response to electrical stimulation or drug application and which correlate with activity recorded in vivo during particular patterns of behaviour. Of these, theta-rhythm (i.e. rhythmical oscillations at 4-12 Hz) has received considerable attention because of its proposed involvement in mnemonic processing. Particularly important to this pattern of activity is the contribution made by the cholinergic system. Indeed, a cholinergic agonist-induced ‘theta-rhythm-like’ activity can promote bidirectional long-term plasticity of glutamate-mediated synaptic transmission (Huerta & Lisman, 1995), a synaptic process often believed to be fundamental for learning. Theta-frequency oscillations in vitro take many forms ranging from intrinsic autorhythmic oscillatory processes at the dendritic (Paulsen & Vida, 1996) and cellular (Leung & Yim, 1991; Strata, 1998) levels to network oscillations generated by the synaptically coupled networks (MacVicar & Tse, 1989). Although it is notable that many of these oscillatory processes in isolated hippocampal preparations appear to be specifically tuned to the theta-frequency range, such processes are distinct from, and should not be confused with, the hippocampal ‘theta-rhythm’ EEG pattern recorded in conscious and anaesthetized animals. Indeed, not all processes are likely to be physiological and an alternative perspective on certain agonist-induced theta-rhythm-like activities is that they may represent a type of epileptiform bursting activity (Williams & Kauer, 1997). Whatever the case, synchronization of hippocampal neuronal networks has important consequences for CNS function and a comprehensive understanding of the mechanisms operating to co-ordinate the activity of populations of neurones in this area of the brain is of particular interest. In this respect, carbachol-induced oscillations have been studied using a variety of approaches and, whilst some findings have been contradictory (MacVicar & Tse, 1989; Traub et al. 1992; Williams & Kauer, 1997), the general consensus is that a close integration of cholinergic, glutamatergic and GABAergic synaptic inputs is paramount to the generation of network synchronization. However, in concentrating on cholinergic aspects, most studies have focused on muscarinic acetylcholine receptor (mAChR)-mediated mechanisms, overlooking the contribution made by nicotinic acetylcholine receptors (nAChRs) to the patterning of rhythmical network activity. Here we specifically address this issue.

METHODS

Slice preparation

Female Wistar rats (2-8 weeks old) or Normal mice (3-5 weeks old) were cervically dislocated and subsequently decapitated in accordance with UK Home Office guidelines. The brain was removed rapidly and transverse parasaggital hippocampal slices were prepared as described previously (Morton & Davies, 1997) by hemisecting the whole brain minus the cerebellum and cutting 400 μm thick transverse brain slices containing hippocampal slices using a vibroslicer (Campden Instruments, Loughborough, UK). Hippocampal slices were cut free from the surrounding brain areas using a scalpel blade and the resultant slices were placed on a nylon mesh at the interface of a warmed (32-34°C), perfusing (1-2 ml min−1) artificial cerebrospinal fluid and an oxygen-enriched (95 % O2, 5 % CO2), humidified atmosphere. The standard perfusion medium comprised (mM): NaCl, 124; KCl, 3; NaHCO3, 26; NaH2PO4, 1.25; CaCl2, 2; MgSO4, 1; D-glucose, 10; and was bubbled with 95 % O2-5 % CO2.

Electrophysiology

Following a 1 h equilibration period intracellular current clamp recordings were obtained from CA3 and/or CA1 pyramidal cells using 2 M potassium methylsulphate-filled microelectrodes (60-110 MΩ). Extracellular recordings were made from stratum pyramidale in area CA3 using microelectrodes (1-5 MΩ) filled with 2-3 M NaCl. Data were captured directly onto DAT tape (DTR-1404; Biologic Scientific Instruments, Claix, France) and/or onto a PC hard disk using AxoScope 1.1 (Axon Instruments, Inc.). In all experiments drugs were applied by addition to the perfusion medium or by drop application in area CA3 via pressure ejection from blunt microelectrodes. With the exception of ondansetron (courtesy of Dr P. Larkman, University of Edinburgh), all other drugs were purchased from commercial sources (Sigma and Tocris Cookson, Bristol, UK). Data are presented as means ± standard error of the mean (s.e.m.) and statistical significance was determined using Student's unpaired t tests or Mann-Whitney U test performed on raw data with P < 0.05 being taken as indicating statistical significance; n values refer to the number of times a particular experiment was repeated. Most experiments were performed using hippocampal slices obtained from rats and, although identical responses were also observed in mouse slices, the majority of traces (except those in Figs 1C and 2) and all quantification of data stated refer to experiments performed on slices obtained from Wistar rats.

Figure 1. Characteristics of the carbachol-induced neuronal network activity.

Figure 1

Open in a new tab

In A, upper traces are excerpts from an intracellular current clamp recording from a CA3 pyramidal neurone in the presence of 1 μM carbachol (a), 10 μM carbachol (b) and in the combined presence of 10 μM carbachol and 2 μM NBQX (c) to illustrate burst-mode activity, theta-mode activity and the slow membrane potential oscillation, respectively. The record shown directly below each trace represents the region indicated by an asterisk in the corresponding upper trace, displayed on an expanded time scale. B shows two consecutive theta-episodes recorded in another neurone at -55 and -77 mV, to illustrate that this activity is due to a bombardment of rhythmic EPSPs. In C, drop application of 10 μM carbachol, at the time indicated by the filled triangle and close to the recorded neurone, generates theta-mode activity without prior activation of burst-mode activity. Scale bars: 50 mV, 50 s (A, upper traces); 50 mV, 400 ms (A, lower traces); 50 mV, 2.5 s (B); 50 mV, 40 s (C).

Figure 2. Cholinergic oscillations occur synchronously in hippocampal pyramidal cells.

Figure 2

Open in a new tab

A shows simultaneous recordings from a CA1 (upper trace) and a CA3 (lower trace) pyramidal neurone on a slow time base during a 10 μM carbachol-induced oscillatory state. B shows the middle portion of the trace in A on an expanded time base and illustrates the synchrony of activity between the two hippocampal subfields. Scale bars: 100 mV, 25 s (A); 100 mV, 1 s (B).

Fimbria-fornix lesions

Uni- or bilateral fimbria-fornix transections were performed on 3-week-old male and female Wistar rats under avertin (tribromoethanol; 10 ml kg−1) anaesthesia. In sham-operated animals the surgical procedure was identical except that no cut was performed. The level of anaesthesia was monitored by assessing eyelid and paw withdrawal reflexes. Surgical procedures were short and did not require supplementary doses of anaesthetic. Animals were placed in a stereotaxic frame with the head horizontal. The skin was removed to expose the skull and small holes were drilled in the skull (1 mm posterior to Bregma) to form a cut 2 mm laterally over both hemispheres. For bilateral cuts, a scalpel blade was lowered into the right hemisphere (-4.5 mm deep from skull) and moved laterally to the same co-ordinates on the contralateral side to ensure transection of all fimbria-fornix fibres. All surgical equipment was soaked in an antiseptic and the wound was cleaned with ethanol. Following surgery, the wound was sutured and the antibiotic powder (Aureomycin) applied. For the first few hours of recovery, all animals were placed in a clean, pre-warmed group cage where they were monitored closely. After this time they were transferred to a clean home cage where they had free access to food and water in a light- and temperature-controlled room. No infection or abnormal behaviour was noticed in any of the rats at any time.

Immunocytochemistry

Animals were killed 2-3.5 weeks after surgery as described above and cholinergic denervation was verified by immunostaining for the vesicular acetylcholine transporter (vAChT) in immersion-fixed (4 % paraformaldehyde, 0.05 % glutaraldehyde and 0.2 % picric acid in 0.1 M phosphate buffer, pH 7.4) slices which had been cryoprotected (1 h in 10 % sucrose, then 1-2 h in 30 % sucrose) and cut into 30 μm sections. Prior to overnight incubation with goat anti-vAChT (1 :1000; Chemicon International, Temecula, CA, USA), sections were blocked with 10 % normal donkey serum. After a 2 h incubation in biotinylated donkey anti-goat IgG, and several washes in Tris-buffered saline (0.05 M, pH 7.4), sections were incubated in avidin-biotinylated HRP (ABC 1 : 1000; Vector Laboratories) for 1.5 h. Immunoreactivity was visualized using 3,3′-diaminobenzidine (DAB) as a substrate. Sections were washed in distilled water before being mounted onto gelatin-coated slides.

In order to obtain tissue with good structural morphology for illustrative purposes (Fig. 7), vAChT immunocytochemistry was also carried out on perfusion-fixed material. Two unilaterally lesioned rats were deeply anaesthetized with pentobarbitol (Sagital, 150 mg kg−1) and then perfused through the heart with saline (1 min) followed by fixative solution (as above, for 10-12 min). Following careful removal, brains were placed in the same fixative solution for a further 1-2 h before being cryoprotected overnight in 30 % sucrose, and 30-50 μm sections were cut in the coronal plane using a freezing microtome (Leica, UK). Following an identical immunocytochemical protocol to that described above, sections were counterstained with Methyl Green (0.5 % aqueous solution) before being scrutinized at the light microscopic level with low- and high-powered objectives to assess the relative density of vAChT-immunoreactive terminals in the control versus the fimbria-fornix-lesioned hippocampus. The lesions were judged to be successful when all but occasional patches of immunoreactive terminals were observed in the hippocampus of the lesioned side of the brain when compared with the dense innervation of vAChT-positive terminals observed in the unlesioned (control) hippocampus (Matthews et al. 1987).

Figure 7. Fimbria-fornix lesions do not prevent theta-mode activity.

Figure 7

Open in a new tab

A, vertical scatter plot showing the characteristics of theta-mode activity in slices prepared from rats following fimbria-fornix transection (triangles represent individual experiments) and sham surgery (circles). No significant difference was observed between groups. B, immunostaining using anti-vAChT antibody as a marker for cholinergic terminals confirming a substantial depletion of cholinergic terminals in the right (lesioned) hippocampal formation compared with the left (unlesioned) side of the brain, 2 weeks following unilateral fimbria-fornix transection. In C and D, the areas indicated by rectangles in the micrograph in B are shown at higher power (C, unlesioned; D, lesioned). Note the relative paucity of vAChT-positive terminals (arrowhead) in the lesioned side (D). Scale bars: 200 μm (B); 20 μm (C and D).

RESULTS

Characterization of cholinergic network activity

Bath perfusion of the AChR agonist carbachol induced an initial depolarization, which often resembled a plateau potential (Fraser & MacVicar, 1996), followed by the appearance of spontaneous synaptic activity in CA3 pyramidal neurones.

‘Burst-mode’, as described here, predominated when low concentrations of carbachol (0.1-1 μM) were used and comprised large individual depolarizing events that occurred at low frequency (< 1 Hz) and which produced synchronous pyramidal cell discharge (n = 63; Fig. 1Aa). Higher concentrations of agonist (10-60 μM) generated ‘theta-mode’ behaviour in which periodic episodes of rhythmic oscillatory depolarization (4-12 Hz) occurred (n = 57; Fig. 1Ab and B). This intermittent activity was coincident with the depolarized phase of a slow intrinsic oscillation (Fig. 1Ab and Ac) and partially resembled rhythmical slow activity or theta-rhythm recorded in vivo (MacVicar & Tse, 1989; Williams & Kauer, 1997). Drop application of 10 μM carbachol enabled theta-mode activity to be generated without prior expression of burst-mode (n = 6; Fig. 1C). Once either mode of activity was established it persisted only in the continued presence of carbachol (n = 63).

In agreement with previous studies, the α-amino-3-hydroxy-5-methylisoxazole-4-propionate (AMPA) receptor antagonist 2,3-dihydroxy-6-nitro-7-sulphamoyl-benzene (f)quinoxaline (NBQX; 1-4 μM; n = 6; Fig. 1Ac), but not the N-methyl-D-aspartate (NMDA) receptor antagonist D-aminophosphonovalerate (AP5; 50 μM; n = 3; not illustrated), abolished theta-mode and burst-mode activity illustrating that both comprised AMPA receptor-mediated synaptic potentials. These activities were coherent across a large population of cells as shown by extracellular recordings as well as simultaneous intracellular recordings from pairs of pyramidal cells (n = 16; Fig 2 and Fig 3B). Each mode of activity was generated in the CA3 region and propagated to other areas since neither was generated in area CA1 when this was severed from area CA3 (n = 7; not illustrated).

Figure 3. Effect of selective nAChR and mAChR ligands on cellular and extracellular field activity.

Figure 3

Open in a new tab

A, traces showing the effect of selective nAChR (a; 10 μM nicotine) and mAChR (b; 10 μM oxotremorine-M) agonists on a CA3 pyramidal cell membrane potential. Note that nicotine has no detectable effect whereas addition of oxotremorine-M to the same cell induces a cholinergic oscillatory state. Ba, simultaneous intracellular (IC) and extracellular (EC) field potential recordings from CA3 pyramidal neurones. Initially both recordings exhibit 10 μM carbachol-induced theta-mode activity which is synchronized, as shown by the traces displayed on an expanded time scale in Bb. Following the addition of 100 μM tubocurarine, at the time indicated by the filled triangle, theta-mode is switched to burst-mode activity which remains coherent between the two recordings despite becoming dissociated from the slow intracellular membrane potential oscillation (Bc). Scale bars: 50 mV, 25 s (A); 50 mV, 200 s (Ba, IC trace); 0.5 mV, 200 s (Ba, EC trace); 50 mV, 25 s (Bb and Bc, upper traces); 0.5 mV, 25 s (Bb and Bc, lower traces).

Role of nAChRs and mAChRs in cholinergic network activity

Since carbachol activates both mAChRs and nAChRs, experiments were performed to address the respective roles of each type of receptor in the initiation, patterning and maintenance of the network activity. The selective mAChR antagonist atropine (5 μM) abolished both burst- and theta-mode activity (n = 6; not illustrated). In contrast, the nAChR antagonist tubocurarine (100-300 μM) did not affect burst-mode activity per se but switched the CA3 network from theta-mode to burst-mode activity (n = 4; Fig. 3B).

At the cellular level, this was accompanied by an increase in the period of an intrinsic membrane potential oscillation (Fig. 3Ba and Bc, upper traces) and its dissociation from the network bursting activity (Fig. 3Ba and Bc, lower traces). Conceptually, these data point to (1) mAChRs being involved in the initiation/maintenance and (2) involvement of nAChRs in the patterning of carbachol-induced network activity. To address these hypotheses directly we assessed the effect of the selective nAChR and mAChR agonists nicotine and oxotremorine-M, respectively. In six slices, nicotine (10-100 μM) failed to induce either pattern of activity (Fig. 3Aa). In contrast, oxotremorine-M, at low concentrations (0.1-1.0 μM), readily induced burst-mode activity that could be converted into theta-mode activity by increasing its concentration (5-50 μM; n = 105; Fig. 3Ab). Analysis revealed that oxotremorine-M- and carbachol-induced theta episodes were similar. Thus, the respective durations of oxotremorine-M- and carbachol-induced theta episodes were 3.87 ± 0.28 and 6.39 ± 2.12 s and occurred every 20.8 ± 1.7 and 25.0 ± 6.3 s. Within each theta episode there were 31 ± 4 and 48 ± 25 oscillatory events occurring at a mean frequency of 8.00 ± 0.65 and 7.86 ± 3.38 Hz, respectively.

Surprisingly, tubocurarine (100 μM) disrupted oxotremorine-M-induced theta-mode activity in the same manner as that observed for carbachol-induced theta-mode activity (n = 15).

Tubocurarine is known to display some affinity for other ligand-gated ion channels including GABAA and 5HT3 receptors. In light of this, we additionally tested a range of nicotinic ligands to confirm the role of nicotinic receptors in the patterning of hippocampal activity. High concentrations of nicotine (100-300 μM; n = 10; Fig. 4A), which should occupy/desensitize nAChRs, as well as nAChR antagonists such as mecamylamine (100-500 μM; n = 8 of 11; Fig. 4B) and dihydro-β-erythroidine (250 μM; n = 4 of 4; Fig. 4C) also switched oxotremorine-M-induced theta-mode activity into burst-mode behaviour. In contrast, the nAChR α7-subunit-preferring antagonist methyllycaconitine (0.1-1 μM; n = 7) had no effect on ongoing theta-mode activity suggesting that activation of α7-subunit-containing nAChRs was not critically involved (Fig. 4D). Finally, to confirm that part of the action of tubocurarine was not mediated through antagonism of 5HT3 receptors, the 5HT3 receptor antagonist ondansetron (2 μM, n = 3; not shown) was applied, with no discernible effect on theta-mode activity.

Figure 4. nAChR ligands switch theta-mode into burst-mode activity.

Figure 4

Open in a new tab

A-C illustrates that nicotine (300 μM), mecamylamine (500 μM) and dihydro-β-erythroidine (250 μM), respectively, switch 10 μM oxotremorine-M-induced theta-mode extracellular activity (left-hand traces; Control) into burst-mode activity, whereas in D, 1 μM methyllycaconitine does not. Scale bars: 0.5 mV, 15 s (A); 0.5 mV, 25 s (B-D).

Does the nicotinic acetylcholine-mediated effect act through GABAergic circuits?

Whilst the presence of functional nAChR-mediated postsynaptic responses in pyramidal cells is controversial (Alkondon et al. 1997; Jones & Yakel, 1997; Frazier et al. 1998a, b; McQuiston & Madison, 1999), GABAergic interneurones have been shown consistently to display both fast and slow nAChR-mediated responses to both locally applied nAChR agonists and cholinergic afferent stimulation (Alkondon et al. 1997; Jones & Yakel, 1997; Frazier et al. 1998a, b; McQuiston & Madison, 1999). To investigate the possibility that the nAChR-mediated switching of network activity described above involves, or may, in some way, be mediated through, GABAergic systems, we compared the effect of the GABAA receptor antagonist bicuculline to that of the nAChR antagonists. Bath application of high concentrations of bicuculline (100 μM) were found to reversibly disassemble periodic episodes of theta-mode activity into burst-mode in a manner identical to that caused by nAChR antagonists (Fig. 5). Moreover, this effect was reversible, with the network activity returning to theta-mode upon washout of the antagonist.

Figure 5. Antagonism of GABAA receptors also switches network activity.

Figure 5

Open in a new tab

In A, the incidence of theta episodes is plotted for the duration of a typical extracellular experiment to examine the role of GABAergic transmission during 10 μM oxotremorine-M-induced theta-mode activity. In B, the upper trace shows a selected portion of the extracellular field recording between 20 and 50 min (inverted triangles), showing the effect of bicuculline application. The lower three traces show 2 min expanded sections of the upper trace before (a), during (b) and after (c) disruption of regular theta-mode activity. Note that application of 100 μM bicuculline reversibly converts regular theta episodes to single bursting events, similar to that produced by 100 μM tubocurarine. Scale bars: 0.25 mV, 3 min (B, top trace); 0.25 mV, 12 s (Ba-Bc).

Origin of nAChR activation

Whilst these data substantiated the respective roles of mAChRs and nAChRs in the initiation/maintenance and temporal patterning of neuronal network activity they also implied that, whereas mAChR activation resulted from the action of exogenously applied agonist, the activation of nAChRs resulted from the release of ACh within the slice. To test this we assessed the ability of hemicholinium-3 and vesamicol, competitive inhibitors of terminal choline uptake and vesicular ACh transporters (vAChTs), respectively, to disrupt oxotremorine-M-induced theta-mode activity. Both agents readily disassembled theta-mode activity such that in the presence of hemicholinium-3 the predominant pattern of network activity was one of a burst-mode interspersed with occasional theta episodes (n = 9; Fig. 6A), whilst in the presence of vesamicol pure burst-mode activity prevailed (n = 14; Fig. 6B). The effects of both compounds were reversible upon washout and, in the case of hemicholinium-3, could be partially reversed by co-application of 10 mM choline (Fig. 6A).

Figure 6. Hemicholinium-3 and vesamicol disrupt theta-mode activity.

Figure 6

Open in a new tab

In A, the graph shows the incidence of theta episodes plotted for the duration of a typical extracellular experiment, to study the effect of hemicholinium-3 on 10 μM oxotremorine-M-induced theta-mode activity. The upper trace (a) shows an example of the theta-mode activity recorded under steady-state control conditions. Following the application of hemicholinium-3, theta-mode activity switches to a predominant burst-mode interspersed with occasional theta episodes (middle trace; b). The substantial delay for this drug effect presumably reflects the time during which existing releasable ACh stores are depleted. Subsequent addition of 10 mM choline reverts network activity to a predominant theta-mode activity interspersed with the occasional burst (bottom trace; c). B, the upper three traces represent an extracellular field potential recording to illustrate that vesamicol reversibly switches oxotremorine-M-induced theta-mode activity into burst-mode activity. The lower traces (a-c) are excerpts from this experiment, shown on an expanded time base, before (a), during (b) and after (c) the effect of vesamicol application, at the corresponding time points indicated on the upper traces. Scale bars: 2 mV, 40 s (Aa-Ac); 2 mV, 100 s (B, upper three traces); 2 mV, 2 s (Ba-Bc).

To explore the origin of the endogenously released ACh lesions of the fimbria-fornix were performed in a group of animals in vivo to degenerate cholinergic terminals arising from the medial septal complex, the primary source of efferent cholinergic fibres to the hippocampal formation.

Slices prepared 2-3 weeks following destruction of the septal cholinergic afferents did not differ from slices prepared from control sham-operated rats, with both groups displaying burst- and theta-mode activity in response to the mAChR agonist oxotremorine-M. Analysis of theta episodes (Fig. 7A) revealed that in sham-operated and lesion groups, mean theta episode durations were 2.86 ± 0.33 and 3.12 ± 0.41 s, respectively, with each episode occurring every 16.5 ± 1.0 and 17.6 ± 2.1 s, respectively. Within theta episodes there were 18 ± 2 oscillatory events occurring at a mean frequency of 6.5 ± 0.6 Hz in sham-operated animals compared with 17 ± 3 events at a frequency of 5.7 ± 0.6 Hz in lesion-operated animals. In both groups theta-mode activity could be disrupted using either tubocurarine or vesamicol (n = 11).

During preparation of slices from fimbria-fornix-lesioned animals, it was noticeable that there was a void where the knife-cut had been made and that the septo-hippocampal projection had been completely severed. However, to further confirm the integrity of the fimbria-fornix lesions and subsequent degeneration of the cholinergic afferents, immunostaining using an antibody against vAChT, as a cholinergic marker, was employed in slices taken from lesioned as well as sham-operated animals. The presence of vAChT-positive terminals was greatly reduced in such lesion-operated animals compared with sham-operated animals. A similar dramatic reduction was also observed in the lesioned hippocampus of rats following unilateral fimbria-fornix lesions, when compared with the intact control side (Fig. 7B-D).

DISCUSSION

Whilst previous investigations have shown that carbachol can induce several distinct forms of synchronized rhythmical activity within the isolated hippocampal slice (Konopacki et al. 1987; MacVicar & Tse, 1989; Heynen & Bilkey, 1991; Bianchi & Wong, 1994; Huerta & Lisman, 1995; Williams & Kauer, 1997; Fisahn et al. 1998; McMahon et al. 1998), only two distinct modes of rhythmical behaviour, burst-mode and theta-mode, were observed here. The apparent absence of other activities presumably results from differences in the strain and age of animals used as well as differences in the slice preparation (see Fisahn et al. 1998). Both burst- and theta-mode activity exhibited a high degree of synchrony and were expressed as synchronous bombardments of AMPA receptor-mediated depolarizing events. In addition, during theta-mode activity there appeared to be an interaction between this phasic efferent excitation and a slow intrinsic membrane potential oscillation (Williams & Kauer, 1997), suggesting that, as is the case for hyperpolarizing synaptic events (Cobb et al. 1995), excitatory synaptic potentials can also interact with intrinsic conductances at the cellular level to provide a powerful means by which hippocampal activity can be synchronized. Given that both burst- and theta-mode activity appear to be generated in the CA3 area, it is tempting to speculate that the strong recurrent glutamatergic excitatory interconnectivity within this region contributes to both. However, the trigger for initiating each mode of activity is the stimulation of mAChRs, the sustained activation of which is necessary to maintain the network activity.

Significantly, however, selective activation of mAChRs is only sufficient to synchronize populations of neurones in burst-mode. To convert this activity to the more complex theta-mode required the recruitment of nAChRs. This finding is supported by the results of Williams & Kauer (1997) and suggests that nAChRs can be viewed as an important mechanism for switching the network from one oscillatory state to another. By selecting a completely distinct modality of neuronal activity, nAChRs therefore directly modify the pattern of synaptic integration within the hippocampus. The finding that activation of this switch occurs through the release of endogenous ACh, and not by the applied agonist, suggests that this is a physiologically relevant mechanism and not a pharmacological artefact. Indeed, the observation that, in vivo, nicotine injected directly into the hippocampus inhibits the EEG theta-rhythm, and also disrupts theta-mode activity in vitro, supports this concept (Ott et al. 1983). Mechanistically, nicotine presumably disrupts these activities through (1) competition with endogenously released ACh for receptor occupancy, (2) its non-phasic nature of application and (3) its capacity to desensitize receptors.

Whatever the case, that nAChR activation underpins the switch from one mode of network activity to another is likely to be reflected in the activation of different cell types and circuits by mAChRs and nAChRs, as has been demonstrated recently in the neocortex (Xiang et al. 1998). However, the nature of the ACh source and location of the target nAChRs within the hippocampal network responsible for patterning neuronal activity are particularly difficult to address. Clearly both are located in area CA3 of the hippocampus since synchronized network activity can be induced in mini-wedges of this hippocampal subfield. Potentially, nAChRs located either pre- and/or postsynaptically may be involved in patterning. Whilst evidence of synaptic activation of nAChRs has remained elusive until recently (Frazier et al. 1998a), several pharmacological studies have demonstrated functional roles for hippocampal nAChRs at both loci. At the presynaptic level, activation of nACh heteroreceptors has been shown to evoke glutamate release from mossy fibre terminals (Gray et al. 1996). However, these are unlikely to be critical for maintaining theta-mode episodes since they are acutely sensitive to the α7-subunit-selective antagonist methyllycaconitine, to which theta-mode activity is completely insensitive. At the postsynaptic level, the presence of functional nAChR-mediated postsynaptic responses in pyramidal cells is subject to much controversy (Alkondon & Albuquerque, 1991; Alkondon et al. 1997; Jones & Yakel, 1997; Frazier et al. 1998a, b; McQuiston & Madison, 1999). However, it certainly remains a possibility that all, or a subpopulation of, pyramidal cells are excited by activation of nAChRs during cholinergic oscillatory states. That said, a stronger candidate population of neurones activated by nAChRs are GABAergic interneurones. Indeed, several subtypes of GABAergic interneurone have been shown to participate in theta-mode behaviour (McMahon et al. 1998). However, it remains unclear whether these cells are instrumental in patterning this network activity or whether they merely receive a similar phasic input to that to pyramidal neurones. Whilst some of these cells express α7-subunits (Alkondon et al. 1997; Jones & Yakel, 1997; Frazier et al. 1998a, b; McQuiston & Madison, 1999), suggesting their non-involvement in theta-mode activity in this study, it is likely that other nAChR subunits are also expressed either in the same neurones or in other interneurones (McQuiston & Madison, 1999). As such, it is conceivable that these cells may fulfil a network-patterning function. Indeed, a recent study by McQuiston & Madison (1999) convincingly demonstrated that different anatomically defined interneurones respond differently to nAChR activation. In particular, some interneurones fail to respond to nAChR agonists, some exhibit responses wholly mediated through α7-subunit-containing receptors and others, whose somata are located in stratum oriens and whose axons mainly ramify within stratum lacunosum-moleculare, exhibit compound nicotinic responses with both fast methyllycaconitine-sensitive and slow dihydro-β-erythroidine-sensitive components.

Whilst it is tempting to speculate that the dihydro-β-erythroidine-sensitive, methyllycaconitine-insensitive α7 theta-mode activity observed in this study involves the latter type of interneurone, further detailed experiments will be required to establish the respective roles of different interneurone subtypes during hippocampal oscillatory states. What is clear is that GABAergic interneurones are involved in some way in theta-mode activity since blockade of fast GABAergic transmission blocks this type of activity.

Whatever the target neuronal population(s) activated by nAChRs (i.e. certain GABAergic interneurones, principal cells or both), given the relatively high calcium permeability of nAChRs, their phasic activation during cholinergic network oscillations may provide an attractive mechanism for coincidence detection and/or providing the conditions necessary for changes in synaptic strength. Such a concept is especially appealing considering that theta-frequency oscillations may also be relevant for the induction of NMDA receptor-dependent synaptic plasticity (Huerta & Lisman, 1995; Paulsen & Moser, 1998).

Another key question arising from these results is where does the ACh that activates the nAChRs originate from? Several possibilities exist including: (1) extrinsic cholinergic afferents and (2) intrinsic hippocampal cholinergic interneurones (Frotscher et al. 1986; Matthews et al. 1987). With respect to the former, the primary population of extrinsic afferents originates from the medial septal complex. Classically, this brain area is considered to provide a phasic drive which mediates hippocampal theta-rhythm in vivo (Stewart & Fox, 1990). However, since the septal cholinergic afferents comprise only axon terminals in the isolated hippocampal slice, it seems unlikely that such a system would act as a phasic source of acetylcholine during theta-mode activity. Moreover, degeneration of these terminals in slices prepared 2-3.5 weeks following lesion of this pathway failed to prevent the network displaying theta-mode behaviour. However, it is conceivable that cholinergic afferents entering the hippocampus by alternative routes (Swanson et al. 1987) may provide sufficient ACh to support this type of activity. Another intriguing possibility is that intrinsic circuits containing cholinergic interneurones provide activation of nAChRs. Interestingly, these cells may project specifically to GABAergic interneurones (Freund & Buzsaki, 1996). The involvement of such a mini-circuit would be consistent with the observation that the GABAA receptor antagonist bicuculline disrupts theta-mode activity in a manner similar to nAChR antagonists, as reported here and confirmed by another laboratory (Williams & Kauer, 1997). Clearly, further anatomical and pharmacological studies will be required to unravel the exact cell types and circuits involved in the various hippocampal oscillatory states.

In summary, therefore, this study demonstrates that activation of nAChRs is critical for switching the hippocampal network between oscillatory states. Such a link between brain nAChRs and cortical oscillatory behaviour may provide important insights into how both are involved in cortical function.

Acknowledgments

We thank Jon Cooper, Ann Wright and Gordon Arbuthnott for their advice on immunocytochemistry. This work was supported by The Wellcome Trust.

References

  1. Alkondon M, Albuquerque EX. Initial characterization of the nicotinic acetylcholine receptors in rat hippocampal neurons. Journal of Receptor Research. 1991;11:1001–1021. doi: 10.3109/10799899109064693. [DOI] [PubMed] [Google Scholar]
  2. Alkondon M, Pereira EFR, Barbosa CTF, Albuquerque EX. Neuronal nicotinic acetylcholine receptor activation modulates γ-aminobutyric acid release from CA1 neurons of rat hippocampal slices. Journal of Pharmacology and Experimental Therapeutics. 1997;283:1396–1411. [PubMed] [Google Scholar]
  3. Bianchi R, Wong KS. Carbachol-induced synchronized rhythmic bursts in CA3 neurons of guinea pig hippocampus in vitro. Journal of Neurophysiology. 1994;72:131–138. doi: 10.1152/jn.1994.72.1.131. [DOI] [PubMed] [Google Scholar]
  4. Buzsáki G, Bragin A, Chrobak JJ, Nadasdy Z, Sík A, Hsu M, Ylinen A. Oscillatory and intermittent synchrony in the hippocampus: relevance to memory trace formation. In: Buzsáki G, Llinas R, Singer W, Berthroz A, Christen Y, editors. Temporal Coding in the Brain. Berlin: G. Springer-Verlag; 1994. pp. 145–172. [Google Scholar]
  5. Buzsáki G, Leung LW, Vanderwolf CH. Cellular bases of hippocampal EEG in the behaving rat. Brain Research Reviews. 1983;6:139–171. doi: 10.1016/0165-0173(83)90037-1. [DOI] [PubMed] [Google Scholar]
  6. Cobb SR, Buhl EH, Halasy K, Paulsen O, Somogyi P. Synchronization of neuronal activity in hippocampus by individual GABAergic interneurons. Nature. 1995;378:75–78. doi: 10.1038/378075a0. [DOI] [PubMed] [Google Scholar]
  7. Fisahn A, Pike FG, Buhl EH, Paulsen O. Cholinergic induction of network oscillations at 40 Hz in the hippocampus in vitro. Nature. 1998;394:186–189. doi: 10.1038/28179. 10.1038/28179. [DOI] [PubMed] [Google Scholar]
  8. Fraser DD, MacVicar BA. Cholinergic-dependent plateau potential in hippocampal CA1 pyramidal neurons. Journal of Neuroscience. 1996;16:4113–4128. doi: 10.1523/JNEUROSCI.16-13-04113.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Frazier CJ, Buhler AV, Weiner JL, Dunwiddie TV. Synaptic potentials mediated via α-bungarotoxin-sensitive nicotinic acetylcholine receptors in rat hippocampal interneurons. Journal of Neuroscience. 1998a;18:8228–8235. doi: 10.1523/JNEUROSCI.18-20-08228.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Frazier CJ, Rollins YD, Breese CR, Leonard S, Freedman R, Dunwiddie TV. Acetylcholine activates an α-bungarotoxin-sensitive nicotinic current in rat hippocampal interneurons, but not pyramidal cells. Journal of Neuroscience. 1998b;18:1187–1195. doi: 10.1523/JNEUROSCI.18-04-01187.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Freund TF, Buzsáki G. Interneurons of the hippocampus. Hippocampus. 1996;6:347–470. doi: 10.1002/(SICI)1098-1063(1996)6:4<347::AID-HIPO1>3.0.CO;2-I. 10.1002/(SICI)1098-1063(1996)6:4<347::AID-HIPO1>3.0.CO;2-I. [DOI] [PubMed] [Google Scholar]
  12. Frotscher M, Schlander M, Léránth C. Cholinergic neurons in the hippocampus: A combined light- and electron-microscopic immunocytochemical study in the rat. Cell and Tissue Research. 1986;246:293–301. doi: 10.1007/BF00215891. [DOI] [PubMed] [Google Scholar]
  13. Gray R, Rajan AS, Radcliffe KA, Yakehiro M, Dani JA. Hippocampal synaptic transmission enhanced by low concentrations of nicotine. Nature. 1996;383:713–716. doi: 10.1038/383713a0. 10.1038/383713a0. [DOI] [PubMed] [Google Scholar]
  14. Heynen AJ, Bilkey DK. Induction of RSA-like oscillations in both the in vitro and in vivo hippocampus. NeuroReport. 1991;2:401–404. doi: 10.1097/00001756-199107000-00012. [DOI] [PubMed] [Google Scholar]
  15. Huerta PT, Lisman JE. Bidirectional synaptic plasticity induced by a single burst during cholinergic theta-oscillation in CA1 in vitro. Neuron. 1995;15:1053–1063. doi: 10.1016/0896-6273(95)90094-2. 10.1016/0896-6273(95)90094-2. [DOI] [PubMed] [Google Scholar]
  16. Jefferys JGR, Traub RD, Whittington MA. Neuronal networks for induced ‘40Hz’ rhythms. Trends in Neurosciences. 1996;19:202–208. doi: 10.1016/s0166-2236(96)10023-0. 10.1016/S0166-2236(96)10023-0. [DOI] [PubMed] [Google Scholar]
  17. Jones S, Yakel JL. Functional nicotinic ACh receptors on interneurones in the rat hippocampus. The Journal of Physiology. 1997;504:603–610. doi: 10.1111/j.1469-7793.1997.603bd.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Konopacki J, MacIver MB, Bland BH, Roth SH. Carbachol-induced EEG ‘theta’ activity in hippocampal brain slices. Brain Research. 1987;405:196–198. doi: 10.1016/0006-8993(87)91009-2. 10.1016/0006-8993(87)91009-2. [DOI] [PubMed] [Google Scholar]
  19. Leung LW, Yim CY. Intrinsic membrane potential oscillations in hippocampal neurons in vitro. Brain Research. 1991;553:261–276. doi: 10.1016/0006-8993(91)90834-i. 10.1016/0006-8993(91)90834-I. [DOI] [PubMed] [Google Scholar]
  20. McMahon LL, Williams JH, Kauer JA. Functionally distinct groups of interneurons identified during rhythmic carbachol oscillations in hippocampus in vitro. Journal of Neuroscience. 1998;18:5640–5651. doi: 10.1523/JNEUROSCI.18-15-05640.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. McQuiston AR, Madison DV. Nicotinic receptor activation excites distinct subtypes of interneurons in the rat hippocampus. Journal of Neuroscience. 1999;19:2887–2896. doi: 10.1523/JNEUROSCI.19-08-02887.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. MacVicar BA, Tse FWY. Local neuronal circuitry underlying cholinergic rhythmical slow activity in CA3 area of rat hippocampal slices. The Journal of Physiology. 1989;417:197–212. doi: 10.1113/jphysiol.1989.sp017797. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Matthews DA, Salvaterra PM, Crawford GD, Houser CR, Vaughn JE. An immunocytochemical study of choline acetyltransferase-containing neurons and axon terminals in normal and partially deafferented hippocampal formation. Brain Research. 1987;402:30–43. doi: 10.1016/0006-8993(87)91044-4. 10.1016/0006-8993(87)91044-4. [DOI] [PubMed] [Google Scholar]
  24. Morton RA, Davies CH. Regulation of muscarinic acetylcholine receptor-mediated synaptic responses by adenosine receptors in the rat hippocampus. The Journal of Physiology. 1997;502:75–79. doi: 10.1111/j.1469-7793.1997.075bl.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Ott T, Malisch R, Destrade C. Nachweis der transmitterspezifischen Moduolation der langsmen rhythmischen Aktivitat des Hippokampus. Biomedica Biochimica Acta. 1983;42:967–979. [PubMed] [Google Scholar]
  26. Paulsen O, Moser EI. A model of hippocampal memory encoding and retrieval: GABAergic control of synaptic plasticity. Trends in Neurosciences. 1998;21:273–278. doi: 10.1016/s0166-2236(97)01205-8. 10.1016/S0166-2236(97)01205-8. [DOI] [PubMed] [Google Scholar]
  27. Paulsen O, Vida I. Sustained dendritic oscillations at theta frequencies elicited in CA1 pyramidal cells in rat hippocampal slices. The Journal of Physiology. 1996;495P:50P. [Google Scholar]
  28. Singer W. Synchronization of cortical activity and its putative role in information processing and learning. Annual Review of Physiology. 1993;55:349–374. doi: 10.1146/annurev.ph.55.030193.002025. 10.1146/annurev.ph.55.030193.002025. [DOI] [PubMed] [Google Scholar]
  29. Stewart M, Fox SE. Do septal neurons pace the hippocampal theta rhythm. Trends in Neurosciences. 1990;13:163–166. doi: 10.1016/0166-2236(90)90040-h. 10.1016/0166-2236(90)90040-H. [DOI] [PubMed] [Google Scholar]
  30. Strata F. Intrinsic oscillations in CA3 hippocampal pyramids: physiological relevance to theta rhythm generation. Hippocampus. 1998;8:666–679. doi: 10.1002/(SICI)1098-1063(1998)8:6<666::AID-HIPO9>3.0.CO;2-A. 10.1002/(SICI)1098-1063(1998)8:6<666::AID-HIPO9>3.0.CO;2-A. [DOI] [PubMed] [Google Scholar]
  31. Swanson LW, Kohler C, Bjorklund A. The limbic region. I: The septohippocampal system. In: Bjorklund A, Hokfelt T, Swanson LW, editors. Handbook of Chemical Neuroanatomy, Integrated Systems of the CNS. Vol. 5. Amsterdam: Elsevier Science Publishers; 1987. pp. 125–277. part 1. [Google Scholar]
  32. Traub RD, Miles R, Buzsáki G. Computer simulation of carbachol-driven rhythmic population oscillations in the CA3 region of the in vitro rat hippocampus. The Journal of Physiology. 1992;451:653–672. doi: 10.1113/jphysiol.1992.sp019184. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Williams JH, Kauer JA. Properties of carbachol-induced oscillatory activity in rat hippocampus. Journal of Neuroscience. 1997;78:2631–2640. doi: 10.1152/jn.1997.78.5.2631. [DOI] [PubMed] [Google Scholar]
  34. Xiang Z, Huguenard JR, Prince DA. Cholinergic switching within neocortical inhibitory networks. Science. 1998;281:985–988. doi: 10.1126/science.281.5379.985. 10.1126/science.281.5379.985. [DOI] [PubMed] [Google Scholar]

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

RESOURCES