| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Related Papers |
1 Department of Neurophysiology, Division of Neuroscience, Medical School, University of Birmingham, Edgbaston B15 2TT, UK
Abstract
Tetanic stimulation of axons terminating in the CA1 region of the hippocampus induces oscillations in the gamma-to-beta frequency band (13–100 Hz) and can induce long-term potentiation (LTP). The rapid pyramidal cell discharge is driven by a mainly GABAA-receptor-mediated slow depolarization and entrained mainly through ephaptic interactions. This study tests whether cellular compartmentalization can explain how cells, despite severely reduced input resistance, can still fire briskly and have IPSPs superimposed on the slow GABAergic depolarization, and whether this behaviour occurs in vivo. Oscillations induced in CA1 in vitro by tetanic stimulation of the stratum radiatum or oriens were analysed using intracellular and multichannel field potentials along the cell axis. Layer-specific effects of focal application of bicuculline indicate that the GABAergic depolarization is concentrated on tetanized dendrites. Current-source density analysis and characteristics of partial spikes indicate that early action potentials are initiated in the proximal nontetanized dendrite but cannot invade the tetanized dendrite, where recurrent EPSPs and evoked IPSPs were largely suppressed. As the oscillation progresses, IPSPs recover and slow the neuronal firing to ß frequencies, with a small subpopulation of neurons continuing to fire at 
frequency. Carbonic anhydrase dependence, threshold intensity, frequency, field strength and spike initiation/propagation of tetanus-evoked oscillations in urethane-anaesthetized rats, validate our observations in vitro, and show that these mechanisms operate in healthy tissue. However, the disrupted electrophysiology of the tetanized dendrites will disable normal information processing, has implications for LTP induction and is likely to play a role in pathological synchronization as found during epileptic discharges.
(Received 1 October 2004;
accepted after revision 2 November 2004;
first published online 4 November 2004)
Corresponding author M. Vreugdenhil: Department of Neurophysiology, Division of Neuroscience, Medical School, University of Birmingham, Edgbaston B15 2TT, UK. Email: m.vreugdenhil{at}bham.ac.uk
Neuronal oscillations in the gamma (: 30–100 Hz) to beta (ß: 14–30 Hz) frequency bands are associated with cognitive functions, e.g. sensory information processing (Singer & Gray, 1995; Haenschel et al. 2000). Large-amplitude neuronal oscillations at -to-ß frequencies can be evoked in the hippocampus by tetanic electrical stimulation of area CA1 (Taira et al. 1997; Whittington et al. 1997a; Bracci et al. 1999) or the dentate gyrus (Poschel et al. 2003) and have been used to assess mechanisms underlying physiologically relevant -to-ß oscillations (Whittington et al. 1997a) and their modulation by psychoactive or anaesthetic drugs (Faulkner et al. 1998; Whittington et al. 2000). Questions have been raised as to whether this is a good model for physiological oscillations (Bracci et al. 1999). The slow depolarization underlying tetanus-induced oscillations has now been shown to be caused by excessive GABAA receptor activation (Kaila et al. 1997; Bracci et al. 1999) and is mediated by bicarbonate efflux through GABAA receptors (Ruusuvuori et al. 2004) due to increased concentrations of intracellular chloride (Isomura et al. 2003) and extracellular potassium (Kaila et al. 1997), partly through the activity of the neuronal potassium-chloride cotransporter KCC2 (Smirnov et al. 1999; DeFazio et al. 2000).
Despite the dramatic loss of input resistance (Kaila et al. 1997) and the sustained depolarization recorded at the soma, pyramidal cells are still able to fire at high frequency (Taira et al. 1997; Bracci et al. 1999). Synchronization of the tetanus-induced firing of pyramidal cells at -to-ß frequencies has been ascribed to both ephaptic interactions (Jefferys, 1995; Bracci et al. 1999) and rhythmic synchronized IPSCs (Whittington et al. 1997a, 2001). The contribution of the latter is impaired by the resistance loss and the depolarized reversal potential of the GABAA receptor-mediated conductance. However, hyperpolarizing IPSPs were recorded superimposed on a slow GABAAergic depolarization in cells recorded some distance from the stimulus electrode (Bracci et al. 1999). Because the CA1 network has a tightly layered structure of excitatory afferents and local inhibitory inputs we hypothesize that tetanic stimulation has differential effects on different cellular compartments, resulting in selective changes in membrane potential, membrane properties and driving force for GABAA receptor-mediated conductances. In this paper we therefore assess the layer-specific oscillatory response of CA1 pyramidal neurons to localized tetanic stimulation.
The oscillations induced by GABAAergic depolarizations (Taira et al. 1997; Bracci et al. 1999; Isomura et al. 2003) and synchronized by ephaptic interaction, observed in our and other laboratories have been attributed to nonphysiological conditions of the brain slices (Whittington et al. 2001). In order to determine whether this was the case, we repeated some of the in vitro experiments in vivo in urethane-anaesthetized rats.
Methods
Slice preparation
Male Sprague-Dawley rats (180–350 g, Harlan OLAC, Bicester, UK) were anaesthetized with by intraperitonal injection of a 7.4 mg kg–1 ketamine–0.7 mg kg–1 medetomidine mixture and killed by cervical dislocation, in accordance with the UK Animals (Scientific Procedures) Act 1986. The brain was removed from the skull, chilled in ice-cold artificial cerebrospinal fluid (aCSF) and cut into 400 µm slices parallel to the midline using a Vibroslice (Campden Instruments, Sileby, UK). The slices were stored in oxygenated aCSF in a storage chamber at room temperature. Slices were transferred to a recording chamber at 33°C and maintained at the interface between warm moist 95% O2–5% CO2 gas mixture and oxygenated aCSF, which consisted of (mM): NaCl 125; KCl 3; NaHCO3 26; NaH2PO4 1.25; CaCl2 2; MgCl2 1; D-glucose 10; pH was equilibrated at 7.4 with 95% O2–5% CO2. The flow rates of moisturized gas (400 ml min–1) and perfusate (3 ml min–1) ensured that the slices were covered by a film of perfusate, matching the control conditions given in Whittington et al. (2001). Except for bicuculline methiodide, drugs were applied to the aCSF in the bath. Halothane was obtained from Zeneca (Macclesfield, UK). All other drugs were obtained from Sigma (Poole, UK).
Bicuculline methiodide (0.2 mM) was applied focally by low-pressure injection into the tissue. A glass pipette with 
10 µm tip diameter was loaded with aCSF containing bicuculline and 0.1 mM Evans blue as a colour indicator. The tip was lowered 0.15 mm into the tissue and gentle pressure (4–7 kPa) was applied while electrical stimulation was applied at 0.1 Hz, until a change in evoked potentials was observed (usually after 20 s). At this point a distinct blue area <0.2 mm diameter was visible. Control experiments used a similar size blue area generated by focal application of aCSF with Evans blue only.
In vitro electrophysiology
Slices were allowed to recover for 1 h before recording. Extracellular field potentials were recorded either with glass pipettes or with an array of eight wire electrodes. Micropipettes were made from 1.2 mm outer diameter borosilicate glass and filled with aCSF (3–5 M
). Field potentials were amplified with Neurolog AC-coupled NL 104 preamplifiers (Digitimer Ltd, Welwyn, UK). Wire electrode arrays were constructed from eight 25 µm 80% nickel–20% chromium wires (Advent Research Materials Ltd, Halesworth, UK) fixed in line, centres 80 µm apart, using dental acrylic (Howmedia International Ltd, London, UK). Electrodes were insulated (except for the tip) by a coat of FORMVAR and had a resistance of <75 . The electrode array was placed in area CA1b perpendicular to the cellular layer at 150 µm from the stimulus electrodes. The electrode array was placed carefully, just touching the tissue, in order to prevent damage. Field potentials were amplified with Digitimer D-169 preamplifiers and D-150 amplifiers (Digitimer Ltd, Welwyn, UK). Intracellular recordings were made from neurons in the stratum pyramidale with sharp (50–80 M) pipettes, filled with 2 M potassium methylsulphate, using an Axoclamp-2A amplifier (Axon Instruments, Burlingham, CA, USA) in bridge mode. Impaled cells had an input resistance ranging from 25 to 55 M and were accepted for recording when the membrane time constant was >10 ms, the resting membrane potential was more negative than –55 mV and stimulation of the stratum radiatum evoked an overshooting action potential. The resting membrane potential was manually adjusted to –65 mV with <0.25 nA command current.
Slices were stimulated with a square pulse of 0.1 ms, using a bipolar electrode made from a pair of twisted FORMVAR-coated 50 µm 80% nickel–20% chromium wires (Advent Research Materials Ltd, Halesworth UK). Stimulus electrodes were placed in the stratum radiatum or in the stratum oriens at 0.15 mm from the stratum pyramidale. Slices were selected that showed clear paired-pulse inhibition with paired stimulation (10 ms interval) of the stratum radiatum. Oscillations were induced in slices by trains of 20 stimuli (tetanus) at 100 Hz (Whittington et al. 1997a). The threshold stimulus intensity was determined for each slice and each stimulation position as the minimum stimulus intensity to elicit small rhythmic population spikes at frequency near the stimulus site. The minimum interval between successive tetani was 3 min.
In vivo electrophysiology
Male Sprague-Dawley rats (230–330 g, Harlan OLAC, Bicester, UK) were anaesthetized with urethane (1.5 mg kg–1 I.P. followed by 1.0 mg kg–1 S.C.) and fixed in a stereotaxic frame. Pure oxygen was supplied near the nostrils at 0.5 l min–1. The body temperature was recorded with a rectal probe and kept constant at 36°C with a heating pad. A window (2 mm diameter) was drilled in the bone above the left dorsal hippocampus. A 16-channel (centres 50 µm apart) silicon recording probe (kindly donated by the Centre for Neural Communication Technology, University of Michigan) was placed at (Pellegrino coordinates from bregma) AP, –2.5 mm; ML, 2.5 mm and lowered into the hippocampus to 3.5 mm from the cortical surface. A silver–silver chloride reference electrode was placed in the contralateral hemisphere in a separate burr hole. Field potentials at eight selected channels were amplified with Digitimer D-169/D-150 amplifiers. A stimulating electrode bundle was made from two pairs of twisted FORMVAR-coated 50 µm 80% nickel–20% chromium wires (Advent Research Materials Ltd, Halesworth UK) with one pair 0.35 mm shorter than the other. The stimulus electrode bundle was placed at AP, –2.7 mm; ML, 2.6 mm and advanced until a single stimulus with the short electrode pair elicited an antidromic population spike. In this way the long electrode pair was reproducibly placed in the stratum radiatum. The depth of the recording probe was adjusted so that the 11th channel from the tip was placed in the stratum pyramidale, where the antidromic population spike (alveus stimulus) was initiated and had its maximal amplitude (Kloosterman et al. 2001). Channels 7–14 were selected for further recording. Oscillations were elicited by tetanic stimulation as described above. At the end of an experiment the animal was killed by cervical dislocation. The work was performed under the UK Animals (Scientific Procedures) Act 1986.
Analysis
The signals were filtered at 3 kHz and sampled at 10 kHz, using a CED 1401 interface and Signal software (Cambridge Electronic Design Ltd, Cambridge, UK). Stimulus artefacts were suppressed offline.
One-dimensional current source density analyses (Mitzdorf & Singer, 1978) were made from eight channel field potential recordings perpendicular to the cellular layer using the following equation:
|
|
x the tissue conductivity at location x. Because tissue conductivity is layer specific and is highest and least variable in the stratum radiatum (Holsheimer, 1987; López-Aguado et al. 2001) we used a value for x recorded in vitro (Holsheimer, 1987) or in vivo (López-Aguado et al. 2001), normalized to its value in the stratum radiatum. We therefore expressed the current source density as mV/mm2, which is proportional to the true current source density. The use-dependent decrease in tissue conductivity with repetitive discharges (Autere et al. 1999; Fox et al. 2004) was not included in the current-source density analysis, because the layer-specificity of use-dependent changes in tissue resistivity under our conditions is unknown. The use-dependent change in resistance under the conditions of Fox et al. (2004), peaked at 25% and was restricted to the stratum pyramidale, which suggests that here we may overestimate the currents in the stratum pyramidale. Data are given as mean ± S.E.M. n indicates the number of slices or cells for in vitro experiments or the number of animals for in vivo experiments. Differences between means were analysed by Student t test with a significance criterion of P < 0.05.
Results
Spatial profile of tetanus-induced oscillations in vitro
Tetanic stimulation (20 pulses at 100 Hz) was delivered at increasing intensities in either the stratum oriens (SO) or stratum radiatum (SR) to determine the threshold for small field potential oscillations, which was 8 ± 1 V for SR-tetani (n = 43) and 10 ± 2 V for SO-tetani (n = 23). Oscillation frequency at threshold stimulus intensity was in the gamma frequency band (; 30–100 Hz): 82 ± 2 Hz for SR-tetani and 79 ± 2 Hz for SO-tetani. The stimulus intensity was fixed at twice-threshold stimulation intensity for experiments in vitro. At this stimulus intensity the frequency of the oscillations gradually reduced from to the ß frequency band (ß: 13–30 Hz, Fig. 1). Pyramidal cells recorded simultaneously close to the stimulus site responded to tetanic stimulation with an initial hyperpolarization (max –5.4 ± 0.3 mV for SR-tetani versus –3.6 ± 0.5 mV for SO-tetani, n = 17, P < 0.05), followed by a slow depolarization (maximum 13.3 ± 0.7 mV at 0.6 ± 0.1 s after the tetanus for SR-tetani versus 15.6 ± 0.6 mV at 0.5 ± 0.1 s after the tetanus for SO-tetani, n.s.). During the depolarization, most cells fired action potentials phase-locked to the population spikes (Fig. 1A and B).
|
In order to study the distribution of the activity over the cell axis the field potential was recorded in eight sites perpendicular to the cell layer at the time of population spike peaks in the stratum pyramidale (SP), for oscillations evoked by tetani in either the SR or SO (Fig. 1C). The maximum population spike amplitude (measured from trough to peak) occurred around the SP and was 8 ± 1 mV (n = 9) for SR-tetani and 7 ± 1 mV for SO-tetani in the same slices. The field potential profile of population spikes during the -phase shows that population spikes are absent in the tetanized layer. It also demonstrates the existence of strong electric fields around the SP with a maximum of 47 ± 6 V m–1 (mean of five largest cycle population spikes in each of nine slices) for SR-tetani and 55 ± 8 V m–1 for SO-tetani. A one-dimensional current-source density analysis (corrected for layer-specific differences in tissue conductivity in vitro, see Methods) was made from the field recordings at the time to peak in the SP. The distribution of the current-source density over the cell axis was dependent on the site of stimulation and indicates that nontetanized layers contain the current sink (negativity, indicating inward currents underlying action potential generation), whereas the tetanized layer serves as a current source (Fig. 1D).
Aberrant firing characteristics
Cell firing in area CA1 is normally initiated in the soma (Turner et al. 1991) or initial segment of the axon (Golding & Spruston, 1998), and the action potential backpropagates into both apical and basal dendrites (Miyakawa & Kato, 1986; Turner et al. 1991; Kloosterman et al. 2001). The site of initiation and the propagation of tetanus-induced discharges were assessed in current-source density–time plots. Figure 2A shows the mean current-source density–time plots for population spikes induced by SO-tetani. For large population spikes during the -phase (Fig. 2Aa) the current sink for the population spike was located in the nontetanized layers with a current source in the SO. Remarkably, the current sink started in the SR, propagated towards the SP and then re-invaded the SR. The maximum backpropagating current sink amplitude in the SR (160 µm from the SP) was 38 ± 5% (n = 9) of that in the SP, a value which is significantly higher than that with a single antidromically evoked population spike recorded in the same slices without a tetanus (17 ± 2%; P < 0.01; see Fig. 2Da).
|
-phase (Fig. 2Ba) the current sink started in the proximal SO, was maximal in the SP, and then returned into the SO, whereas the tetanized layers remained passive. This suggests that during the -phase the action potential is initiated in the nontetanized soma and proximal basal dendrites and fails to backpropagate into the apical dendrite. The inability of the tetanized dendrite to discharge was further assessed for using antidromically evoked discharges. Field potentials were recorded simultaneously from different CA1 layers during an antidromic population spike evoked by stimulation of the alveus (Fig. 2Ca). The mean current-source density analysis- time plot (Fig. 2Da) shows a current sink, starting in the SP and propagating into both the SR and SO (Miyakawa & Kato, 1986; Turner et al. 1991; Kloosterman et al. 2001). When an antidromic population spike was evoked 0.2 s after an SR-tetanus (Fig. 2Cb), no current sink could be observed in the SR, whereas the sink in the SO was increased (Fig. 2Db). This confirms that the tetanized dendrite is unable to discharge.
During the ß-phase, the spike initiation site of large population spikes shifts towards the soma with SO-tetani (Fig. 2Cb) and is confined to the soma with SR-tetani (Figs 2Db).
These data indicate that tetanic stimulation of either the apical or the basal dendritic layer: (i) prevents action potential invasion into the dendrites in the tetanized layer (whether spontaneous or evoked by antidromic stimulus), (ii) promotes action potential invasion in the dendrites in the nontetanized layer, and (iii) temporarily shifts the action potential initiation area during the -phase from the soma towards the proximal nontetanized dendrite.
Miniature population spikes and spikelets
In the majority of slices, small-amplitude population spikes were found throughout the nontetanized layers between the large population spikes during the ß-phase (Fig. 3A) or just in front of large population spikes in the -phase. These miniature population spikes run at frequency (64 ± 2 Hz with SR-tetani (n = 22) and 66 ± 2 Hz with SO-tetani (n = 13). The current-source density analysis of the miniature population spikes during the ß-phase (Fig. 2Ac, and Bc) located the sink in the soma and proximal apical dendrite for both SR-tetani and SO-tetani. If the perisomatic sink is active, these miniature population spikes are likely to reflect the synchronous discharge of a small subpopulation of cells. However, if the perisomatic sink is passive (with an active source in the tetanized layer), they may reflect rhythmic IPSCs resulting from ongoing oscillations in the tetanized interneuron network (Whittington et al. 1997b). Similar interneuron network gamma was first described in the absence of pyramidal cell activity (Whittington et al. 1995). Because interneuron frequency is determined by the IPSC time constant (Whittington et al. 1995), we tested the effect on oscillation frequency of increasing the IPSC time constant with the barbiturate thiopental (20 µM). In order to match the slow depolarization and the oscillation amplitude and duration in control, the stimulus intensity was reduced (by 10–30%). Although thiopental reduced the frequency of large population spikes in the ß-phase (from 19.6 ± 1.5 Hz to 13.4 ± 1.0 Hz, P < 0.001, as previously reported by Faulkner et al. (1999), it did not significantly affect the frequency of miniature population spikes during the ß-phase (from 61.6 ± 1.9 Hz to 63.7 ± 2.3 Hz in thiopental, n = 9, n.s., Fig. 3A). These observations suggest that the ongoing oscillation during the ß-phase reflects the rhythmic firing of a (subcompartment of a) subpopulation of cells that escapes inhibitory control.
|
frequency during the ß-phase, where they coincided with the miniature population spikes (Fig. 3B) and occasionally triggered action potentials (not shown). Spikelets occur most frequently at the peak of the slow depolarization (Fig. 3C). In some cells, the depolarization is so strong that all firing is temporarily disrupted, similarly to that described for low-Ca2+-induced slow depolarizations (Bikson et al. 2003), probably due to depolarization block resulting from sodium channel inactivation. In those cells, spikelets precede and follow the silent period (not shown). In cells with spikelets, most action potentials had a shoulder on the upstroke (inset Fig. 3D), similar to that found with antidromic spike invasion from the axon (Turner et al. 1993). Antidromic ‘spikelets’ in CA1 have been attributed both to invasion of ectopic action potentials generated in synaptic terminals or axons (Stasheff et al. 1993), and to invasion of action potentials generated by electrotonic coupling of axons through gap junctions (Draguhn et al. 1998; Schmitz et al. 2001). However, here we found that the putative gap junction blockers halothane (5 mM, n = 6, Fig. 3C), octanol (0.3 mM, n = 3), 18
-glycyrrhetinic acid (0.1 mM, n = 3) and carbenoxolone (0.2 mM for 60 min, n = 4) and the gap junction opener trimethylamine (10 mM, n = 3) failed to alter the occurrence and appearance of tetanus-induced spikelets during the -phase. In eight neurons, SR-tetani elicited spikelet doublets or triplets with an interspike interval of 1–3 ms (Fig. 3E, see also Fig. 3B, arrow). These spikelet multiplets were never observed in the same cells with SO-tetani (Fig. 3D). With the shortest interspike intervals, temporal summation of spikelets could trigger full-sized action potentials (Fig. 3E). The high-frequency (up to 1 kHz), the variability of interspike interval and the variability of spikelet amplitude within each spikelet multiplet (inset Fig. 3E) are incompatible with firing patterns from a single axon. A more likely explanation for spikelet multiplets is the independent initiation of action potentials within a short period in different basal dendrites that converge on the recorded soma.
Layer-specific phasic synaptic potentials
The layer-specific firing properties (Fig. 2) and the drop in input resistance (Kaila et al. 1997; Bracci et al. 1999)(see Table 1) suggest that the large shunting conductances underlying the slow depolarization are concentrated on the tetanized dendrite. Local interneuron axonal projections are layer specific (Lambert et al. 1991; Freund & Buzsáki, 1996). Although, polysynaptic GABA release is not restricted to the stimulated layer (Freund & Buzsáki, 1996; Fujiwara-Tsukamoto et al. 2004), we therefore predict that phasic synaptic inputs will be less suppressed at the nontetanized dendrite than at the tetanized dendrite.
|
cycles (Fig. 4Aa; Table 1). Test stimuli applied at 0.6 s showed significant, but incomplete, recovery and a longer suppression of the oscillation (Fig. 4Ab; Table 1). A similar recovery of synaptic potentials and interruption of the oscillation was seen with SO stimulation following an SR-tetanus (Table 1). In contrast, postsynaptic potentials evoked by test stimuli applied in the SO following an SO-tetanus (Fig. 4B) or in the SR following an SR-tetanus were small (Table 1) and had little effect on the oscillation.
|
CA1 neurons have axon collaterals with synapses confined to the basal dendrite (Deuchars & Thomson, 1996) and cause rhythmic recurrent EPSPs that gradually increase during tetanus-induced oscillations (Whittington et al. 1997b). In the majority of cells tested SR-tetani caused discernible recurrent EPSPs as the frequency slowed to ß (Fig. 4Ca). In some cells recurrent EPSPs could evoke a late action potential (• in Fig. 4Ca) when the cell did not fire with the population spike. As predicted from a layer-specific shunting, the recurrent EPSPs following SO-tetani (Fig. 4Cb) were much smaller than those following SR-tetani (compare also Fig. 1A with Fig. 1B), despite the activity of a similar population (as judged from the population spike amplitude). In eight cells, recurrent EPSP amplitude at 0.5–0.6 s after the tetanus was 3.0 ± 0.8 mV for SR-tetani and 0.6 ± 0.3 mV for SO-tetani (P < 0.01). This confirms that the tetanized dendrite is unable to sustain synaptic inputs.
Layer-specific slow potentials
Bath application of the GABAA receptor antagonist bicuculline blocked tetanus-induced oscillations and suppressed the underlying slow depolarization in both submerged (Bracci et al. 1999; Isomura et al. 2003) and interface conditions (Taira et al. 1997; Bracci et al. 1999), independent of the site of stimulation (Bracci et al. 1999). Both hyperpolarizing and depolarizing GABAAergic currents are concentrated in the stimulated hippocampal layer (Lambert et al. 1991), consistent with layer-specific projections of interneurons (Freund & Buzsáki, 1996). We now assessed whether the subcellular location of the GABAAergic slow depolarization was stimulation site-specific, by applying bicuculline focally (see Methods) into the tissue in five slices in vitro. The effect of focal bicuculline application was monitored by paired-pulse (10 ms interval) stimulation to the SR until the responses close to the application site changed, when tetanic stimulation was given. Application in the SR reduced Schaffer collateral stimulus-induced paired-pulse inhibition without generating a second (recurrent) population spike (Fig. 5A). Application in the SO did not affect paired-pulse inhibition, but transiently impaired recurrent inhibitory control as judged from the multiple spikes (Fig. 5C). This indicates that the effect of focally applied bicuculline was concentrated in the layer of application.
|
oscillation (mean amplitude of population spikes during the interval between 0.1 s and 0.2 s after the start of the oscillation) was reversibly suppressed to 7 ± 2% of control for SR-tetani (Fig. 5B) and to 5 ± 3% for SO-tetani. Intracellular recordings show that focal application of bicuculline to the stimulated layer prevented the initial hyperpolarization during the tetanus (Figs 5Ea), and reduced the maximum amplitude of the subsequent slow depolarization to 56 ± 13% for SO-tetani in three out of three cells (P < 0.05) and to 54 ± 8% for SR-tetani in three out of three cells (P < 0.05, Figs 5Ea). Focal bicuculline application did not affect the firing pattern or the afterhyperpolarization evoked in three neurons by a 0.2 s 1 nA current injection (data not shown), indicating that effects of bicuculline on potassium conductances (Khawaled et al. 1999) were small under the present conditions.
When bicuculline was applied at the opposite, nonstimulated layer (Fig. 5D) the oscillation amplitude increased slightly to 123 ± 8% (P < 0.05) of control values for SR-tetani and to 115 ± 9% (n.s.) for SO-tetani. The frequency of the early oscillation following SR-tetani did not change (84 ± 3 Hz versus 82 ± 3 Hz in control, n.s., example in Fig. 5D), but the frequency of large population spikes during the ß-phase increased (from 17.8 ± 2.5 Hz to 26.6 ± 2.6 Hz, P < 0.05). The slow depolarization did not decrease when bicuculline was applied in the layer opposite to the stimulation site: it was unchanged for SO-tetani (111 ± 3%; n = 3, n.s., Fig. 5Eb) and enhanced for SR-tetani (121 ± 5%; n = 3, P < 0.05). This modest increase in depolarization can explain the increased oscillation and may result from reduced shunting and/or reduced hyperpolarization of the nontetanized dendrite.
These observations suggest that the GABAAergic depolarization underlying cell firing is concentrated in the tetanized dendrites, and that the GABAAergic conductance in the nontetanized compartments reduces excitability and oscillation frequency during the ß-phase.
Tetanus-induced oscillations in vivo
Whittington et al. (2001) suggested that ephaptic interactions were undetectable under normal conditions, and were only seen under conditions of decreased osmolarity or insufficient perfusion, resulting in dry slice surfaces. Our in vitro observations were performed at normal osmolarity in well-perfused and wetted slices and were similar to observations in the submerged condition (Bracci et al. 1999). However, in order to determine whether or not the phenomenon we describe here could be due to a nonphysiological condition of the slices, we performed multichannel recordings of SR-tetanus-induced oscillations in eight rats under urethane anaesthesia. Using bipolar stimulus electrodes identical to those used in vitro, the threshold intensity for oscillations with 20 pulses in vivo (7 ± 1 V) was not different from that in vitro, but oscillations started before the end of the tetanus. Large (5–10 mV) population spikes were elicited in the SP in all animals tested at stimulus intensities just above (1.1–1.3 times) threshold, confirming in vivo the nonlinearity of the response observed in vitro. Single stimuli at twice-threshold intensity evoked single orthodromic population spikes in five out of eight rats. In these animals a second stimulus delivered 10 ms later never elicited a population spike, indicating strong synaptic inhibition. At this stimulus intensity 4–6 pulses were sufficient to evoke oscillations, whereas 20 pulses caused a reduction in the amplitude of spontaneous population spikes, suggesting that the resulting depolarization was strong enough to cause depolarization block (Bikson et al. 2003). We therefore chose to use 10 pulses for subsequent experiments. At threshold intensity for 10 pulses, rhythmic small population spikes occurred in the SP and SR at 92 ± 3 Hz (range 75–102 Hz; Fig. 6A). Twice-threshold intensity 10-pulse tetani elicited oscillations with population spikes in the SP and SO (Fig. 6B). The maximum population spike amplitude was found in the SP (12 ± 1 mV) and was larger than that in vitro (8 ± 1 mV P < 0.01). The maximum strength of electric fields around the SP (56 ± 13 mV mm–1) was not different from that in vitro. The frequency of large population spikes was 71 ± 3 Hz for the first 10 population spikes, dropping to 11 ± 2 Hz < 0.5 s after the start of the oscillation (Fig. 6C). Although the frequency of large population spikes can drop below the ß frequency band, we call the late phase the ß-phase. As in vitro, small population spikes run at frequency (74 ± 2 Hz, n = 8) between the ß beats (inset Fig. 6C), significantly faster than with SR-tetani in vitro (P < 0.01). The relatively high frequencies of miniature population spikes and threshold oscillations are likely to be caused by the higher temperature in vivo, but are within the range previously observed with tetanus-induced oscillations in CA1 (Bracci et al. 1999) and the dentate gyrus (Poschel et al. 2003) in vitro. The duration of the oscillation was 1.5 ± 0.2 s, significantly shorter than with SR-tetani in vitro (2.3 ± 0.2 s, P < 0.05).
|
-to-ß oscillation in vitro depends on production of bicarbonate by carbonic anhydrase (Staley et al. 1995; Kaila et al. 1997; Bracci et al. 1999; Ruusuvuori et al. 2004). The carbonic anhydrase blocker acetazolamide (Na+ salt, 200 mg kg–1 I.P.; Fig. 6D) reduced the amplitude of the tetanus-induced oscillation in vivo by 87 ± 5% (n = 3), suggesting an important role for bicarbonate production in tetanus-induced activity in vivo. However, we cannot exclude an indirect action of acetazolamide, e.g. through changes in pH.
Judging from the profile of activity over the cell axis in vivo (Fig. 6B), the action potentials failed to invade the tetanized apical dendrites. One-dimensional current-source density analysis (corrected for layer-specific differences in tissue conductivity in vivo) were made from multichannel recordings in eight rats. Figure 7A shows the mean current-source density–time plots for population spikes induced by SR-tetani. The somatic sink for population spikes during the -phase is preceded by an early sink in the nontetanized dendritic layer (Fig. 7Aa) and fails to invade the tetanized layer, just as in vitro. For large population spikes during the ß-phase the spike initiation site shifts to the SP (Fig. 7Ab). The sink related to miniature population spikes during the ß-phase was located perisomatically, just as in vitro. Antidromic population spikes were elicited by single stimuli to the alveus before and 0.2 s after an SR-tetanus (Fig. 7B). The current-source density-time plot of the antidromic population spike showed backpropagation of the discharge into the SR (Fig. 7Ca; averaged from four rats). Just after an SR-tetanus, the current sink of the evoked antidromic population spike was increased in the SP and SO but absent in the SR (Fig. 7Cb), confirming the observation in vitro that the action potential cannot backpropagate into the tetanized dendrite.
|
Discussion
The coherent oscillatory response of CA1 to tetanic stimulation shows remarkable compartmentalization. Depolarizing GABA potentials are concentrated in the tetanized dendrites, which cease to support action potentials or to respond to phasic synaptic input, while the nontetanized dendrites can initiate action potentials and support phasic synaptic inputs. Tetanus-evoked oscillations in vivo had similar properties to those in vitro, suggesting that they were produced by healthy tissue. The aberrant synaptic and firing properties will disable normal information processing and are likely to pave the way for epileptiform activity.
Layer-specific modulation of synaptic activity
The hippocampus CA1 area has a particularly layered organization of afferent inputs (Amaral, 1993) and local interneuron axonal projections (Freund & Buzsáki, 1996). Although, given the variety of interneurons activated by local stimulation, polysynaptic GABA release is not restricted to the stimulated layer (Freund & Buzsáki, 1996; Fujiwara-Tsukamoto et al. 2004), monosynaptic GABAAergic currents are concentrated in the stimulated hippocampal layer (Lambert et al. 1991). The layer-specific effect of focal bicuculline shows that the tetanus-induced GABAA conductance underlying the initial slow depolarization is concentrated on the tetanized dendrites. Excessive GABAAergic activity can depolarize the membrane directly by bicarbonate efflux through GABAA gated anion channels (Kaila, 1994; Isomura et al. 2003; Ruusuvuori et al. 2004) and indirectly by a rise in extracellular potassium concentration (Smirnov et al. 1999) through bicarbonate-dependent interneuron activity, as proposed by Kaila et al. (1997). Given the considerable local shunting, the bicarbonate efflux will dominate the initial depolarization, whereas a potassium rise will gain significance later in the response, when input resistance recovers.
The layer-specific tonic GABAAergic input has a direct impact on phasic synaptic inputs. IPSPs on the tetanized dendrites are strongly suppressed, probably due to localized shunting (Jackson et al. 1999), along with presynaptic suppression of release, postsynaptic receptor desensitization and rise in intracellular chloride concentration. Although the latter is limited at the soma by the large volume : surface ratio (Staley & Proctor, 1999), somatic IPSCs are likely to be attenuated by dendritic shunting. Phasic hyperpolarizing IPSPs on the nontetanized dendrites show partial recovery with time, consistent with a relatively high local membrane resistance and relatively normal intracellular chloride levels, which keep the GABAA equilibrium potential more negative than the local membrane potential. This compartmentalization of the response to tetanic stimulation explains why recordings from the soma can show GABAAergic IPSPs superimposed on a slow GABAAergic depolarization. A use-dependent compartment-specific reversal of IPSCs was reported to contribute to tetanus-induced seizure-like events (Fujiwara-Tsukamoto et al. 2004). The compartment-specific shunting is also demonstrated by the recurrent EPSPs, which are located on the basal dendrite (Deuchars & Thomson, 1996) and consequently much more suppressed by tetanus-induced shunting of the basal dendrite than of the apical dendrite. Recurrent EPSPs recovered from the initial shunting after an SR-tetanus are relatively strong (Bracci et al. 2001) and can now even excite neurons and contribute to synchronizing SR-tetanus-induced oscillations and probably to the initiation of seizure-like events following oscillations (Köhling et al. 2000). A use-dependent increase in recurrent excitation (Whittington et al. 1997b) can be important in epilepsy, because excitatory synaptic activity and local connectivity of CA1 pyramidal cells are increased in epileptic rats (Shao & Dudek, 2004).
Layer-specific disruption of firing properties
In pyramidal neurons, action potentials are normally initiated in the axon initial segment (Colbert & Johnston, 1996), the soma (Turner et al. 1991; Kloosterman et al. 2001) or the stimulated dendrite (Miyakawa & Kato, 1986; Turner et al. 1991), and then propagate into both apical and basal dendrites (Turner et al. 1991). During tetanus-induced fast oscillations, action potential-related current sinks were absent from the tetanized layers, presumably because the tetanized dendrites were unable to fire due to GABAergic shunting (Bracci et al. 1999) and depolarization-induced sodium channel inactivation (Lian et al. 2003; Bikson et al. 2003). In contrast, action potential invasion in the nontetanized dendrites increased, perhaps as a result of the inactivation of A-type potassium current by moderate depolarization (Hoffman et al. 1997).
The early current sink in the nontetanized layer during oscillations, and the frequent occurrence of spikelets suggests that tetanic stimulation shifts the spike initiation site temporarily towards nontetanized cellular compartments. The possibility that this is the axon, perhaps amplified by gap junction coupling (Stasheff et al. 1993; Draguhn et al. 1998; Schmitz et al. 2001) seems unlikely because of: (i) the early sink found in the SR with SO-tetani, (ii) the resistance of tetanus-induced spikelets to gap junction modulators, and (iii) the SR-tetanus-induced spikelet multiplet characteristics. It is more likely that tetanus-induced spikelets are the passive somatic expression of dendritic spikes (Turner et al. 1993; Golding & Spruston, 1998; Kamondi et al. 1998), initiated in the proximal nontetanized dendrite when the soma is unable to fire due to shunting and depolarization block (Bikson et al. 2003; Lian et al. 2003). The spikelet multiplets found with SR-tetani would then reflect action potential initiation in several basal dendrites within a short period. Recent observations confirm the ability of the basal dendrite to initiate discharges (Ariav et al. 2003).
Ephaptic interactions, enhanced by post-tetanic increases in tissue resistance (Autere et al. 1999), play a major role in synchronizing cell firing during tetanus-induced oscillations (Bracci et al. 1999) as well as during ictal-like events in low extracellular calcium (Haas & Jefferys, 1984; Traub et al. 1985; Fox et al. 2004). DC fields are effective in modulating neuronal firing in a layered structure like CA1 (Jefferys, 1995; Bikson et al. 2003, 2004). Although AC fields are less effective, due to membrane capacitance, the field strength of the large population spikes is sufficient to cause (Taylor & Dudek, 1984) and to synchronize the firing of sufficiently excited neurons (Haas & Jefferys, 1984; Traub et al. 1985). The large-amplitude population spikes at just-suprathreshold stimulus intensities in vivo indicates that the majority of pyramidal neurons are depolarized close to firing threshold. The similar field strength in vivo and in vitro suggests that ephaptic interactions play a significant role in synchronization in vivo (Bracci et al. 1999).
The tetanus-induced slow depolarization causes cells to fire initially at high frequencies, similar to those observed in epileptiform burst in the absence of synaptic transmission (Fox et al. 2004). This high-frequency firing is resistant to bicuculline applied to nontetanized layers, and is probably determined by intrinsic properties (Bracci et al. 1999). The gradual increase in interval between large population spikes is associated with the recovery of recurrent IPSPs with time after the tetanus (Bracci et al. 1999) This is confirmed by the gradual, but partial, recovery of IPSPs evoked by stimulation in the nontetanized layer, the parallel increase in ability to delay the next large population spike, and by the modulation of the ß-intervals by focal bicuculline and by thiopental. This suggests that, in the majority of neurons late in the oscillation, the spike-initiating cell compartment is effectively controlled by these GABAAergic synapses, delaying imminent firing. We propose that these IPSPs play a key role in setting the period of the ß-phase, while either a subset or a subcompartment of the neurons continues firing at an intrinsically set (Bracci et al. 1999) frequency, giving rise to miniature population spikes. This was confirmed by the increase in the number of cycles skipped by the majority of neurons during the ß-phase under conditions favouring strong recurrent inhibition, e.g. in vivo and with thiopental in vitro.
Experimental conditions
The present experiments on tetanic oscillations in vivo revealed: a low stimulus threshold; a requirement for carbonic anhydrase activity; high initial oscillation frequency; and layer-specific changes in firing properties. These closely match our in vitro observations under both interface and submerged conditions (Bracci et al. 1999), and concur with observations from other labs (Staley et al. 1995; Taira et al. 1997; Bracci et al. 1999; Isomura et al. 2003), showing that these results cannot be attributed to nonphysiological slice conditions (Whittington et al. 2001). Discrepancies between the present observations in vitro and in vivo and those reported by Whittington et al. (1997a, 2001) could perhaps be explained by a relatively intact GABAAergic inhibition under our in vitro conditions, as judged from the initial hyperpolarization, which is absent in recordings from pyramidal neurons in the work of Whittington et al. (1997a, 2001) and Traub et al. (1999), and when the GABAA conductance is suppressed, e.g. Fig. 5E. Under these conditions, stronger stimulation may depolarize cells via metabotropic glutamate receptors (Whittington et al. 1997a, 2001). GABAergic transmission in vivo, in contrast, is strong and sufficient to cause GABAAergic depolarizations.
Functional consequences
Tetanic stimulation induces aberrant behaviour in pyramidal neurons, both in vitro and in vivo, because the tetanized dendrite is unable to sustain discharges and synaptic inputs but the nontetanized compartments can initiate discharges at frequency and can respond to synaptic inputs. During tetanus-induced oscillations, normal information processing through area CA1 will be severely distorted, and the use of tetanus-induced oscillation to study physiologically relevant oscillations and their role in cognition is unwarranted.
Tetanic stimulation is widely used in long-term potentiation (LTP) induction and relies on backpropagating action potentials (Magee & Johnston, 1997). Under our conditions, the layer-specific GABAAergic shunting and possible inactivation of sodium channels prevents action potential backpropagation into the tetanized dendrite, which may have spatial implications for LTP induction. Indeed, LTP is prevented only in distal dendritic locations when backpropagation is impaired (Isomura & Kato, 2000) and suppression of GABAergic inhibition improves tetanus-induced LTP induction on the tetanized dendrite (Fujii et al. 2000).
Tetanus-induced oscillations display strong similarities with discharges encountered in acute epilepsy models, where depolarizing GABA plays a role (Avoli et al. 1993; Köhling et al. 2000; Isomura et al. 2003) and/or ephaptic interactions are crucial (Haas & Jefferys, 1984; Traub et al. 1985; Traynelis & Dingledine, 1988). Abnormal amplitude oscillations are often observed at seizure onsets in partial epilepsies (Fisher et al. 1992), and spontaneous oscillations, closely resembling tetanus-induced oscillations, can trigger long-lasting seizure-like events (Köhling et al. 2000; Isomura et al. 2003; Fujiwara-Tsukamoto et al. 2004), possibly by an oscillation-induced increase in recurrent excitation (Bracci et al. 2001). Tetanus-induced oscillations therefore provide a model for the study of pathologically relevant hypersynchronization.
References
Amaral DG (1993). Emerging principles of intrinsic hippocampal organization. Curr Opin Neurobiol 3, 225–229.[CrossRef][Medline]
Ariav G, Polsky A & Schiller J (2003). Submillisecond precision of the input-output transformation function mediated by fast sodium dendritic spikes in basal dendrites of CA1 pyramidal neurons. J Neurosci 23, 7750–7758.
Autere AM, Lamsa K, Kaila K & Taira T (1999). Synaptic activation of GABAA receptors induces neuronal uptake of Ca2+ in adult rat hippocampal slices. J Neurophysiol 81, 811–816.
Avoli M, Psarropoulou C, Tancredi V & Fueta Y (1993). On the synchronous activity induced by 4-aminopyridine in the CA3 subfield of juvenile rat hippocampus. J Neurophysiol 70, 1018–1029.
Bikson M, Hahn PJ, Fox JE & Jefferys JG (2003). Depolarization block of neurons during maintenance of electrographic seizures. J Neurophysiol 90, 2402–2408.
Bikson M, Inoue M, Akiyama H, Deans JK, Fox JE, Miyakawa H & Jefferys JGR (2004). Effects of uniform extracellular DC electric fields on excitability in rat hippocampal slices in vitro. J Physiol 557, 175–190.
Bracci E, Vreugdenhil M, Hack SP & Jefferys JGR (1999). On the synchronising mechanisms of tetanically-induced hippocampal oscillations. J Neurosci 19, 8104–8113.
Bracci E, Vreugdenhil M, Hack SP & Jefferys JGR (2001). Dynamic modulation of excitation and inhibition during stimulation at gamma and beta frequencies in the CA1 hippocampal region. J Neurophysiol 85, 2412–2422.
Colbert CM & Johnston D (1996). Axonal action-potential initiation and Na+ channel densities in the soma and axon initial segment of subicular pyramidal neurons. J Neurosci 16, 6676–6686.
DeFazio RA, Keros S, Quick MW & Hablitz JJ (2000). Potassium-coupled chloride cotransport controls intracellular chloride in rat neocortical pyramidal neurons. J Neurosci 20, 8069–8076.
Deuchars J & Thomson AM (1996). CA1 pyramid–pyramid connections in rat hippocampus in vitro: dual intracellular recordings with biocytin filling. Neuroscience 74, 1009–1018.[Medline]
Draguhn A, Traub RD, Schmitz D & Jefferys JGR (1998). Electrical coupling underlies high-frequency oscillations in the hippocampus in vitro. Nature 394, 189–192.[CrossRef][Medline]
Faulkner HJ, Traub RD & Whittington MA (1998). Disruption of synchronous gamma oscillations in the rat hippocampal slice: a common mechanism of anaesthetic drug action. Br J Pharmacol 125, 483–492.[CrossRef][Medline]
Faulkner HJ, Traub RD & Whittington MA (1999). Anaesthetic/amnesic agents disrupt beta frequency oscillations associated with potentiation of excitatory synaptic potentials in the rat hippocampal slice. Br J Pharmacol 128, 1813–1825.[CrossRef][Medline]
Fisher RS, Webber WR, Lesser RP, Arroyo S & Uematsu S (1992). High-frequency EEG activity at the start of seizures. J Clin Neurophysiol 9, 441–448.[Medline]
Fox JE, Bikson M & Jefferys JGR (2004). Tissue resistance changes and the profile of synchronized neuronal activity during ictal events in the low-calcium model of epilepsy. J Neurophysiol 92, 181–188.
Freund TF & Buzsáki G (1996). Interneurons in the hippocampus. Hippocampus 6, 347–470.[CrossRef][Medline]
Fujii S, Jia Y, Yang A & Sumikawa K (2000). Nicotine reverses GABAergic inhibition of long-term potentiation induction in the hippocampal CA1 region. Brain Res 863, 259–265.[CrossRef][Medline]
Fujiwara-Tsukamoto Y, Isomura Y, Kaneda K & Takada M (2004). Synaptic interactions between pyramidal cells and interneurone subtypes during seizure-like activity in the rat hippocampus. J Physiol 557, 961–979.
Golding NL & Spruston N (1998). Dendritic sodium spikes are variable triggers of axonal action potentials in hippocampal CA1 pyramidal neurons. Neuron 21, 1189–1200.[CrossRef][Medline]
Haas HL & Jefferys JGR (1984). Low-calcium field burst discharges of CA1 pyramidal neurones in rat hippocampal slices. J Physiol 354, 185–201.
Haenschel C, Baldeweg T, Croft RJ, Whittington MA & Gruzelier J (2000). Gamma and beta frequency oscillations in response to novel auditory stimuli: a comparison of human electroencephalogram (EEG) data with in vitro models. Proc Natl Acad Sci U S A 97, 7645–7650.
Hoffman DA, Magee JC, Colbert CM & Johnston D (1997). K+ channel regulation of signal propagation in dendrites of hippocampal pyramidal neurons. Nature 387, 869–875.[CrossRef][Medline]
Holsheimer J (1987). electrical conductivity of the hippocampal CA1 layers and application to current-source-density analysis. Exp Brain Res 67, 402–410.[Medline]
Isomura Y & Kato N (2000). Possible regulatory role of dendritic spikes in induction of long-term potentiation at hippocampal schaffer collateral-CA1 synapses. Brain Res 883, 119–124.[CrossRef][Medline]
Isomura Y, Sugimoto M, Fujiwara-Tsukamoto Y, Yamamoto-Muraki S, Yamada J & Fukuda A (2003). Synaptically activated Cl– accumulation responsible for depolarizing GABAergic responses in mature hippocampal neurons. J Neurophysiol 90, 2751–2756.
Jackson MF, Esplin B & Capek R (1999). Inhibitory nature of tiagabine-augmented GABAA receptor-mediated depolarizing responses in hippocampal pyramidal cells. J Neurophysiol 81, 1192–1198.
Jefferys JGR (1995). Non-synaptic modulation of neuronal activity in the brain: electric currents and extracellular ions. Physiol Rev 75, 689–723.
Kaila K (1994). Ionic basis of GABAA receptor channel function in the nervous system. Prog Neurobiol 42, 489–537.[CrossRef][Medline]
Kaila K, Lamsa K, Smirnov S, Taira T & Voipio J (1997). Long-lasting GABA-mediated depolarization evoked by high-frequency stimulation in pyramidal neurons of rat hippocampal slice is attributable to a network-driven, bicarbonate-dependent K+ transient. J Neurosci 17, 7662–7672.
Kamondi A, Acsady L & Buzsáki G (1998). Dendritic spikes are enhanced by cooperative network activity in the intact hippocampus. J Neurosci 18, 3919–3928.
Khawaled R, Bruening-Wright A, Adelman JP & Maylie J (1999). Bicuculline block of small-conductance calcium-activated potassium channels. Pflugers Arch 438, 314–321.[CrossRef][Medline]
Kloosterman F, Peloquin P & Leung LS (2001). Apical and basal orthodromic population spikes in hippocampal CA1 in vivo show different origins and patterns of propagation. J Neurophysiol 86, 2435–2444.
Köhling R, Vreugdenhil M, Bracci E & Jefferys JGR (2000). Ictal epileptiform activity is facilitated by hippocampal GABAA receptor-mediated oscillations. J Neurosci 20, 6800–6829.
Lambert NA, Borroni AM, Grover LM & Teyler TJ (1991). Hyperpolarizing and depolarizing GABAA receptor-mediated dendritic inhibition in area CA1 of the rat hippocampus. J Neurophysiol 66, 1538–1548.
Lian J, Bikson M, Sciortino C, Stacey WC & Durand DM (2003). Local suppression of epileptiform activity by electrical stimulation in rat hippocampus in vitro. J Physiol (Lond) 547, 427–434.
López-Aguado L, Ibarz JM & Herreras O (2001). Activity-dependent changes of tissue resistivity in the CA1 region in vivo are layer-specific: modulation of evoked potentials. Neuroscience 108, 249–262.[CrossRef][Medline]
Magee JC & Johnston D (1997). A synaptically controlled, associative signal for Hebbian plasticity in hippocampal neurons. Science 275, 209–213.
Mitzdorf U & Singer W (1978). Prominent excitatory pathways in the cat visual cortex (A17 and A18): a current source density analysis of electrically evoked potentials. Exp Brain Res 33, 371–394.[Medline]
Miyakawa H & Kato H (1986). Active properties of dendritic membrane examined by current source density analysis in hippocampal CA1 pyramidal neurons. Brain Res 399, 303–309.[CrossRef][Medline]
Poschel B, Heinemann U & Draguhn A (2003). High frequency oscillations in the dentate gyrus of rat hippocampal slices induced by tetanic stimulation. Brain Res 959, 320–327.[CrossRef][Medline]
Ruusuvuori E, Li H, Huttu K, Palva JM, Smirnov S, Rivera C, Kaila K & Voipio J (2004). Carbonic anhydrase isoform VII acts as a molecular switch in the development of synchronous g
, dotted line) and by SO-tetani (
, continuous line). Error bars indicate S.E.M. Paired t test significance levels are indicated as *, P < 0.05; **, P < 0.01; ***, P < 0.001. Arrows indicate the site of tetanic stimulation. D, one-dimensional current source density analysis at the time of the population spike peak in the SP. The tetanized dendrite provided the main source for the population spike-related sink. Symbols as in C.
