INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Interneurons account for 10-12% of the total hippocampal cell population and are responsible for GABAergic inhibition, which controls the excitability of hippocampal neural circuits (Kandel et al. 1961; Nicoll et al. 1990). Furthermore, networks of GABAergic interneurons are responsible for cortical rhythmic activity such as synchronous  (Whittington et al. 1995) and  (Cobb et al. 1995; Tóth et al. 1997) oscillations. Membrane properties of different subclasses of interneurons in stratum (st.) oriens (Lacaille and Williams 1990; Lacaille et al. 1987; McBain 1994), st. pyramidale (Buhl et al. 1994, 1996; Gulyás et al. 1993; Knowles and Schwartzkroin 1981; Miles et al. 1996), st. lacunosum-moleculare (L-M) (Khazipov et al. 1995; Lacaille and Schwartzkroin 1988), and at the border between st. radiatum and L-M (Kawaguchi and Hama 1987; Kunkel et al. 1988; Williams et al. 1994) have been investigated. In particular, interneurons such as basket and axoaxonic cells in st. pyramidale and vertical cells in st. oriens generate short action potentials followed by fast afterhyperpolarizations (AHPs) responsible for sustained high-frequency firing (Buhl et al. 1994, 1996; Lacaille and Williams 1990; Lacaille et al. 1987); conversely, stellate cells in st. L-M possess longer action potentials (followed by a fast AHP) and fire at a slower rate (Lacaille and Schwartzkroin 1988; Williams et al. 1994).

A heterogeneous population of interneurons is known to be present in the hippocampal st. radiatum (Maccaferri and McBain 1996; McMahon and Kauer 1997). These cells contain glutamate decarboxylase (GAD), the enzyme synthesizing the neurotransmitter GABA (Frotscher et al. 1984; Ribak et al. 1978; Woodson et al. 1989) and are therefore presumably inhibitory. Surprisingly, although there are reports concerning the morphology (Gulyás et al. 1993; McMahon and Kauer 1997) and synaptic input (Laezza et al. 1999; Maccaferri and McBain 1996; McBain and Dingledine 1993; McMahon and Kauer 1997) of these cells, a detailed study of the basic electrophysiological properties of interneurons located in this region is still lacking.

Potassium currents involved in the action potential repolarization have been studied in interneurons of st. oriens-alveus (Zhang and McBain 1995), in parvalbumin-positive interneurons in st. pyramidale (Du et al. 1996), and in st. L-M (Chapman and Lacaille 1999). Little is known about the potassium conductances underlying action potential AHPs in the various subclasses of interneurons present in the hippocampus. In st. oriens interneurons, different Ca²⁺-dependent K⁺ conductances have been suggested to underlie an iberiotoxin-sensitive fast AHP (fAHP) and a medium-duration AHP, sensitive to apamin and carbachol (Zhang and McBain 1995). To date, the only data concerning the K⁺ currents underlying the AHP, recorded in the voltage-clamp mode, are related to the recent report on st. L-M interneurons by Aoki and Baraban (2000).

The aim of the present work was to investigate the intrinsic membrane properties of st. radiatum interneurons in the CA3 hippocampal region. By combining electrophysiological techniques, pharmacological tests and in situ hybridization, we characterized the properties of the currents underlying the AHP. Potassium currents sensitive to apamin and carbachol were found to contribute to the AHP, playing a fundamental role in shaping the firing pattern.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Slice preparation

Hippocampal slices were prepared from the brain of young (10- to 21-day-old) Wistar rats as previously described (Edwards et al. 1989). In brief, rats were decapitated under anesthesia (5% urethan ip), and their brains were rapidly removed and placed in ice-cold artificial cerebrospinal fluid (ACSF) of the following composition (in mM): 126 NaCl, 3.5 KCl, 2 CaCl₂, 1.2 NaH₂PO₄, 1.3 MgCl₂, 25 NaHCO₃, and 25 glucose, gassed with 95% O₂-5% CO₂ (pH 7.3). After bisecting the brain, the tissue was immersed into low temperature (2-4°C), oxygenated ACSF solution. Transverse slices (250 µm thick) were cut with a vibrating microslicer (Vibracut, FTB, Weinheim, Germany) and incubated for 1 h at 32°C before use.

Whole cell recordings

Tight-seal whole cell recordings were obtained from neurons with cell bodies located in st. radiatum of the CA3b region (Lorente de Nó 1934) at room temperature (22-24°C). Large neurons with a triangular shape were excluded from the sample of interneurons, because of their presumed excitatory origin (Gulyás et al. 1998). Patch pipettes had resistances of 3-4 M and were filled with the following pipette solution (in mM): 130 K-gluconate, 10 KCl, 10 N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES), 0.4 ethylene glycol-bis(-aminoethyl ether)-N,N,N',N'-tetraacetic acid (EGTA), 1 MgCl₂, 0.3 Na GTP, and 2 Na₂ATP (pH 7.3). In current-clamp recordings, voltage responses were recorded in bridge mode with an Axoclamp 2B amplifier (Axon Instruments, low-pass filtering at 10 kHz). Capacitance neutralization and resistance compensation were applied by monitoring the voltage response onset to a current step at high time resolution. In voltage clamp, currents were recorded with an EPC-7 List amplifier, sampled every 0.2 ms and filtered at 2 kHz. All tail potassium currents were recorded in TTX (1 µM) and TEA (0.5 mM) to block voltage-dependent sodium currents and fast potassium currents (BK channels) (Coetzee et al. 1999), respectively.

Histology

Biocytin (0.2%, Sigma) was added to the pipette solution just before use. Injected cells were visualized using the avidin-biotinylated horseradish peroxidase complex reaction (ABC; Vector Laboratories, Burlingame, CA) with cobalt and nickel-intensified 3,3'-diaminobenzidine as chromogen. After slice embedding in glycerol, neurons were viewed under light microscopy. Microphotographs of neurons were taken at different focal depths. After scanning each photo, the pictures were overlaid to reconstruct the whole neuronal morphology observable in a 250 µm-thick slice.

In situ hybridization

In situ hybridization was performed on brain sections (10-16 µm) from male rats using ³⁵S-labeled antisense and sense oligonucleotide probes according to the procedure described in detail (Stocker and Pedarzani 2000). Every SK channel probe was tested on sections obtained from four different animals, two as sagittal and two as coronal cuts. For each SK channel subunit, at least two antisense oligonucleotides corresponding to the 5' and 3' regions with no significant similarity to the other known SK channel subunits were chosen (SK1: 5'-GGCCTGCAGCTCCGACACCACCTCATATGCGATGCTCTGTGCCTT-3' and 5'-CAGTGGCTTT-GTGGGCTCTGGGCGGCTGTGGTCAGGTGACTGGGC-3'; SK2: 5'-AGCGCCAGGTTGTT-AGAATTGTTGTGCTCCGGCTTAGACACCACG-3' and 5'-CTTCTTTTTGCTGGACTTAGTGC-CGCTGCTGCTGCCATGC- CCGCT-3'; SK3: 5'-CGATGAGCAGGGGCAGGGAATTGAAGCTGG-CTGTGAGGTGCTCCA-3' and 5'-TAGCGTTGGGGTGATGGAGCAGAGTCTGGTGGGCATG-GTTATCCT-3'). The sense oligonucleotides had complementary sequences to the second oligonucleotide listed for each channel and were used to control for general background on adjacent sections. Specificity of the observed signals was confirmed in three ways: 1) identical hybridization patterns were obtained with each pair of antisense oligonucleotides; 2) hybridization with a mixture of the same labeled and nonlabeled oligonucleotide in 100- to 500-fold excess did not result in detectable signal (Fig. 5E); 3) a mixture of labeled oligonucleotide specific for a certain SK subunit and nonlabeled oligonucleotides for other SK subunits resulted in the same hybridization pattern as obtained with the specific antisense oligonucleotide alone. For cellular resolution, selected slides were dipped in photographic emulsion (Kodak NTB2) and developed after 12-20 wk. Brain structures were identified according to Paxinos and Watson (1986).

Data analysis

Input resistance (R_in) was estimated from the apparently linear portion of the steady-state voltage-current relationships obtained by measuring the amplitude of voltage responses (15 mV) to hyperpolarizing current pulses. The slower membrane time constant (₀) was determined by fitting either single or double exponential functions to the average of five membrane responses (10 mV) obtained by hyperpolarizing current injection (150 ms). At holding potential ranging from 52 to 54 mV (close to the resting potential) single action potentials were elicited by 3 ms depolarizing pulses (at spike threshold intensity). A single spike just at the end of current injection was evoked to avoid interference by the capacitative transient. Averages of three action potentials are shown in the figures. Different parameters were used to characterize single action potentials: firing threshold, amplitude (measured from the threshold to the peak), and duration (measured at the half-amplitude). mAHP amplitude was measured from the peak to the baseline. The maximum steady-state firing rate was obtained by averaging the instantaneous frequency for the last five intervals of a train evoked by the largest current before onset of spike failure. Accommodation rate was quantified by measuring the instantaneous discharge frequencies between the first two spikes (f_initial), 150 ms after the beginning of the discharge (f₁₅₀), and at the end of the stimulation (f_final) in a train of spikes after injection of 100 pA for 500 ms. Early and late accommodations were calculated according to (f_initial  f₁₅₀)/f_initial and (f₁₅₀  f_final)/f_initial, respectively (Cauli et al. 1997). To quantify inward rectification, the membrane potential was measured at the peak and at the end of a 500 ms hyperpolarizing current injection that brought the cell to 100 mV, and the corresponding rectification ratio was calculated.

All potential values obtained in current-clamp mode were not corrected for the measured liquid junction potential (10 mV), and therefore the values of membrane potential should be considered 10 mV more negative.

Potassium currents underlying the mAHP were studied in voltage clamp. To avoid problems in voltage control, they were studied as tail currents elicited from a holding potential of 50 mV after a depolarizing step (100 ms) of 60 mV (see, for example, Pedarzani and Storm 1993). The time-dependent decay of the tail currents was fitted between 80 and 20% of peak using single or bi-exponential functions. pCLAMP (Axon Instruments) software was used to perform voltage and current acquisition and analysis. Statistical parameters were assessed with Student's t-tests applied to the raw data (P < 0.05). Unless otherwise mentioned, all results are presented as means ± SD.

Drugs

3,3'-Diaminobenzidine (DAB), tetraethylammonium chloride (TEA), carbachol, and CsCl were purchased from Sigma (St. Louis, MO); apamin from Latoxan (Valence, France); and 6,7-dinitroquinoxaline-2,3-dione (DNQX) from Tocris (Bristol, UK) and tetrodotoxin (TTX) from Affinity Research Product (Exeter, UK). All other chemicals were from Merck. Drugs were applied via bath superfusion using a three-way tap system.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Morphology

Fifteen interneurons [postnatal day 10 to 12 (P10-12)] and seven interneurons P17-21 with cell bodies localized in st. radiatum of the CA3b region were morphologically identified with biocytin injection and also investigated electrophysiologically. The boundary between st. radiatum and L-M was distinguished due to the darker appearance of st. L-M (Lacaille and Schwartzkroin 1988). Interneurons showed variable soma shapes (round, stellate, fusiform, or ovoid, see Fig. 1, A and B) of small size (approximately 10-12 µm in maximum diameter). Most of them were multipolar (Fig. 1B) with variable patterns of dendritic arborization, often restricted to st. radiatum. Due to the multipolar orientation, their axons were difficult to see, and they were often cut during slicing. Whenever identifiable, axons were confined to st. radiatum (n = 5) or directed toward the CA3 pyramidal layer (n = 2). Varicosities were often observed along the axons in st. radiatum. Dye coupling was observed in three slices (obtained from different rats at P10-12): in one case involving three cells (Fig. 1C); in the others, two cells. Dye-coupled interneurons showed various somata shapes and dendritic arborizations. We never observed dye coupling in older (P17-21) st. radiatum interneurons (n = 7).

View larger version (86K): [in this window] [in a new window]   Fig. 1. Morphology of stratum (st.) radiatum interneurons as revealed after biocytin injection. A and B: representative samples of electrophysiologically identified CA3 st. radiatum interneurons. The neurons were stained by biocytin (0.2%) added to the patch-pipette solution. Note the multipolar soma in B. Calibration bars, 25 µM. C: 3 dye-coupled CA3 st. radiatum interneurons are labeled following injection of biocytin into 1 cell (see arrow) at postnatal day 10 (P10). Note the variable shapes and multipolar somata of the cells. Dendritic processes radiate into st. radiatum. Calibration bar, 50 µm.

Membrane properties

Current-clamp recordings were obtained from a total of 53 interneurons located in the CA3b region of st. radiatum. Despite the heterogeneous morphology of interneurons, membrane properties appeared to be similar and, for this reason, were pooled together. Little spontaneous synaptic activity and no spontaneous firing were present at P10-21. Basic membrane properties of interneurons were measured at resting membrane potential. They differed from those found in CA3 pyramidal cells and did not show significant changes with postnatal development between P10 and P21 (Table 1). A representative sample of an action potential and its afterpotentials elicited by a brief current pulse is shown in Fig. 2A.

View larger version (22K): [in this window] [in a new window]   Fig. 2. Action potential and firing properties of st. radiatum interneurons. A: sample of a single action potential elicited by a short (3 ms) depolarizing current pulse. B: trains of action potentials elicited by depolarizing current injections of increasing intensity (from left to right: 40 and 100 pA, each 500 ms). C: regular firing is present at all current intensities (marked on the right) with early accommodation only. Numbers on abscissa represent the sequence of action potentials. D: voltage responses to depolarizing and hyperpolarizing current injection. Negative current injection reveals a prominent "sag" typical of activation of the hyperpolarization-activated cation conductance (Ih). A depolarizing rebound triggering action potential follows the turning off of the hyperpolarizing pulse.

Afterpotentials

Single action potentials were followed by a mAHP, which appeared as a direct continuation of the spike repolarization (Fig. 2A). Mean amplitude and duration of mAHP following a single spike were 6.5 ± 2.1 mV and 232 ± 69 ms, respectively (mean ± SD, n = 29). A slow AHP (sAHP, lasting more than 1 s) (Sah 1996) was never observed following single action potentials in st. radiatum interneurons at room temperature (n = 50). In only 8% of interneurons (4 of 50), a sAHP was identified after trains of action potentials.

Repetitive firing and inward rectification

Figure 2B shows representative samples of the firing discharge of a st. radiatum interneuron on injection of increasing depolarizing current pulses starting from the resting potential. The firing pattern was found to be regular for all suprathreshold stimulus intensities (n = 29). Injection of the same depolarizing current pulses when the membrane was hyperpolarized (20 mV from rest) preserved the regularity of firing but slowed down the firing rate and delayed the onset of the first spike (n = 5, not shown). The maximum steady-state firing rate was 31 ± 4 Hz (n = 20), as reported for neocortical regularly spiking interneurons (Erisir et al. 1999). The presence of spike frequency accommodation was investigated in 20 interneurons by injecting 500-ms-long depolarizing current pulses (40-120 pA). In Fig. 2C a plot of the spike sequence versus the reciprocal of the inter-spike interval (instantaneous frequency) for various injected currents shows that the number of spikes increased with the current strength, and that there was relatively weak accommodation. The average of early accommodation was 33 ± 9.8% and occurred mainly over the first few spikes. Late accommodation was minimal (9.4 ± 3.6%, n = 20). Figure 2D shows the voltage response elicited by depolarizing and hyperpolarizing pulses. Negative current injection reveals an inward rectification manifested as a large "sag" typically generated by the hyperpolarization-activated cation conductance (I_h) (Halliwell and Adams 1982). A depolarizing rebound that triggered action potentials followed the turning off of the hyperpolarizing pulse. These phenomena were typically observed in all st. radiatum interneurons investigated in the present report and were both abolished by CsCl (2 mM, n = 3, data not shown). A lower rectification ratio was found in pyramids, in which the depolarizing rebound never triggered a spike.

Potassium conductances underlying mAHP

Several studies indicate that potassium conductances underlying the AHP play a fundamental role in shaping the discharge pattern of various neurons (for a review, see Sah 1996). In the present study the ionic conductances involved in generating the AHP were investigated using various pharmacological tools.

The membrane currents underlying the mAHP were revealed as tail currents under voltage-clamp mode following depolarizing voltage steps to +10 mV (100 ms long) from a holding potential of 50 mV. Representative tail currents recorded in TTX (1 µM) and TEA (0.5 mM) are shown in Fig. 3, A and Ca (control). Figure 3Ba shows the progressive activation of tail currents following 100-ms-long depolarizing steps of increasing amplitude (from 40 to +20 mV, 10-mV increments) starting from a holding potential of 50 mV. The apparent threshold for activation of tail currents was 30 mV (n = 4). Depolarizing voltage steps to +10 mV followed by a final voltage step varying between 75 and 45 mV evoked tail currents with a reversal potential around 75 mV. As the external K⁺ concentration was raised to 6.5 mM, the reversal potential shifted by approximately 10 mV (Fig. 3Bb).

View larger version (22K): [in this window] [in a new window]   Fig. 3. Tail currents are Ca2+ dependent, and sIAHP is present in a small number of cells. A: superimposed tail currents following depolarizing voltage step (100 ms to +10 mV) recorded from a holding potential of 50 mV before and after superfusion of Ca2+-free solution. A partial recovery is obtained. Ba: average current-voltage (I-V) relation (n = 4) of tail currents. Tail currents were normalized with respect to that obtained by stepping the voltage from 50 to +20 mV. Bars are SE. Bb: plot of the average of tail currents evoked in 5 cells either in 3 or 6.5 mM K+ medium by applying 100-ms voltage steps to +10 mV and varying the subsequent voltage step from 75 to 45 mV. Tail current peak amplitudes were measured in respect to the steady-state value reached on decay to baseline and were normalized with respect to the current obtained by stepping the final voltage step from +10 to 45 mV in 3 mM K+ medium. Bars are SE. Ca: superimposed tail currents recorded in the absence and presence of carbachol (40 µM). Cb: same traces as in Ca, but expanded on the amplitude axis: note the presence of a slow outward current completely abolished after superfusion of carbachol.

A slow outward current (lasting more than 1 s), reminiscent of sI_AHP, appeared in 5 of 47 neurons (11%) and was completely and reversibly blocked by Ca²⁺-free solution (n = 2, not shown) or carbachol (40 µM, n = 4, Fig. 3Cb). We cannot exclude that sI_AHP might have been underestimated in this study due to the EGTA-containing pipette solution (see, for example, Madison and Nicoll 1984; but also Zhang et al. 1995) and the low temperature (Lancaster and Adams 1986; Sah 1996).

Apamin-sensitive I_AHP

To investigate whether a calcium-sensitive component takes part in generating the potassium tail currents elicited by short depolarizing pulses, we applied Ca²⁺-free medium containing 1 mM EGTA. After 5 min superfusion of Ca²⁺-free solution, a reversible decrease (58 ± 6%) in the peak amplitude of the tail potassium current was observed (n = 5, Fig. 3A). Superfusion of Cd²⁺ (200 µM) + Ni²⁺ (50 µM) similarly reduced the tail currents (49 ± 6%, n = 6; not shown).

Apamin is known to block specifically Ca²⁺-activated potassium channels of the SK type in different brain regions (Sah 1996). Sensitivity to apamin was tested by recording tail currents in the absence and in the presence of apamin. The proportional contribution of apamin-sensitive tail currents varied appreciably from cell to cell. On average, the effect of apamin was dose dependent, with small reductions in the peak of tail currents observed already at subnanomolar concentrations (see the example in Fig. 4A at 500 pM), and stronger reductions observed at the saturating concentration of 100 nM (Fig. 4Ba). In control conditions, tail currents had a peak amplitude of 206 ± 84 pA and could be fitted by two exponentials with decay time constants () of 15 ± 1 ms and 56 ± 16 ms, respectively (n = 15). As shown in Fig. 4Ba, in the presence of apamin (100 nM) the peak amplitude of the tail currents was depressed by 38.5 ± 17% (range in different cells: from 17 to 79%, n = 15). After digital current subtraction (Fig. 4Bb), the apamin-sensitive tail currents were shown to have a monoexponential decay ( = 50 ± 20 ms, n = 15), corresponding to the slower component of the tail current deactivation. By delivering voltage steps of fixed amplitude (to +10 mV) and increasing duration (from 3 to 100 ms) in the absence and presence of apamin (100 nM), interestingly we observed that even voltage steps as short as 3 ms were able to generate a substantial fraction of apamin-sensitive current (data not shown).

View larger version (12K): [in this window] [in a new window]   Fig. 4. Apamin-sensitive tail currents. A: sample of tail currents following depolarizing voltage steps (100 ms to +10 mV) recorded from a holding potential of 50 mV before and after superfusion of apamin 500 pM. Note that this low dose of apamin is already able to reduce the outward tail currents. Ba: superimposed tail currents recorded from a holding potential of 50 mV before and after superfusion of apamin (100 nM). Bb: apamin-sensitive current obtained by subtracting the records shown in Ba.

Expression of SK channel subunit transcripts in st. radiatum interneurons

Three members of the SK family of Ca²⁺-activated K⁺ channels have been so far cloned and characterized. They give origin to channels with different sensitivity to apamin, the most sensitive being SK2 homomultimers (IC₅₀ ~ 63 pM) (Ishii et al. 1997; Köhler et al. 1996), while SK3 homomultimeric channels present an intermediate sensitivity (IC₅₀ ~ 2 nM: Ishii et al. 1997), and SK1 channels a low sensitivity (IC₅₀ > 100 nM: Ishii et al. 1997; Köhler et al. 1996; IC₅₀ ~ 3.3-12 nM: Shah and Haylett 2000; Strobaek et al. 2000).

To identify molecularly SK channels in CA3 st. radiatum interneurons, we studied the expression of SK subunit mRNAs by in situ hybridization with oligonucleotide probes. Analysis at cellular resolution revealed the presence of only one SK transcript, namely SK2, at high levels in scattered CA3 st. radiatum interneurons (Fig. 5B), whereas SK1 and SK3 mRNAs were below the threshold limit of detection (Fig. 5, A and C). In the adjacent CA3 pyramidal layer, SK subunits presented a different expression pattern, with SK2 being the most abundant transcript (Fig. 5B) but SK1 and SK3 mRNAs displaying also moderate to high expression levels (Fig. 5, A and C) (see also Stocker et al. 1999). The high expression level of SK2 transcript is additionally illustrated in the high power photomicrograph showing clusters of silver grains on single CA3 interneurons (Fig. 5D). These results are in agreement with the presence of an apamin-sensitive Ca²⁺-dependent K⁺ current in CA3 st. radiatum interneurons. Based on the exclusive high expression of SK2 mRNA, our results suggest that homomultimeric SK2 channels might mediate the apamin-sensitive I_AHP in these neurons.

View larger version (140K): [in this window] [in a new window]   Fig. 5. Expression of SK channel subunit transcripts in CA3 st. radiatum interneurons. Dark-field photomicrographs of rat brain coronal sections hybridized with oligonucleotides specific for SK1 (A), SK2 (B), and SK3 (C) subunit mRNAs. E: control. White arrowheads in B indicate high levels of SK2 transcript in CA3 st. radiatum interneurons. Expression of SK1 (A), SK2 (B), and SK3 (C) transcripts is evident in CA3 pyramidal neurons (CA3). D: bright-field high-power photomicrograph showing the high expression level of SK2 subunit mRNA in single CA3 st. radiatum interneurons. Scale bars: A, B, C, and E: 100 µm; D: 40 µm.

Ca²+-dependent apamin-sensitive mAHP

The functional role of the apamin-sensitive Ca²⁺-dependent potassium current was studied in current-clamp mode in Ca²⁺-free solutions as well as in the presence of apamin. A change in action potential waveform was observed after 5-min superfusion with Ca²⁺-free solution, and recovered when standard external solution was reapplied (n = 5). The action potential firing threshold decreased (from 38 ± 2.2 mV to 40 ± 2.5 mV), and a broadening of the action potential at the late stage of repolarization was observed (Fig. 6Aa). Action potential amplitude and repolarizing rate decreased in Ca²⁺-free solutions by 5.3 ± 3.6% and by 25 ± 5.0%, respectively. Action potential duration was prolonged by 30 ± 10%. In three of five cells the mAHP was completely blocked; in the remaining two cells it was reduced by 62%. Reduction or block of mAHP always increased the firing rate (34 ± 14% change, n = 5; see Fig. 6Ab).

View larger version (19K): [in this window] [in a new window]   Fig. 6. An apamin- and calcium-sensitive conductance mediates the medium-duration afterhyperpolarizing potential (mAHP) and regulates the firing frequency. A: superimposed action potentials elicited by brief current pulses before and after 5 min superfusion with Ca2+-free solution. Ca2+-free medium reduces the mAHP (a) and enhances firing frequency (b). B: superimposed action potentials before and after application of apamin (100 nM). Apamin reduces the amplitude of mAHP without affecting action potential waveform (a). The block of mAHP induces an increase in the firing rate during trains of action potentials (b).

Apamin (100 nM) specifically and irreversibly reduced the mAHP amplitude (40 ± 16%) and duration (42 ± 9.2%), without affecting the action potential (Fig. 6Ba, n = 6). In four of six cells the effect on the mAHP was sufficient to significantly increase the rate of action potential firing (17 ± 10% change, Fig. 6Bb), while in the remaining two no significant change in spike frequency was observed.

Carbachol-sensitive potassium conductances

The hippocampus is innervated by cholinergic fibers (Freund and Buzsáki 1996; Miettinen and Freund 1992), and the acetylcholine receptor agonist carbachol is known to affect the mAHP of CA1 pyramidal cells by blocking the potassium current I_M (Storm 1989). In CA3 st. radiatum interneurons the proportional contribute of carbachol-sensitive tail currents in the presence of TTX (1 µM) and TEA (0.5 mM) varied from 14 to 24% of the total tail current (18 ± 4%, n = 6; data not shown). Apamin-insensitive potassium tail currents, recorded in the presence of apamin (100 nM), were further reduced by carbachol (40 µM). A representative example is shown in Fig. 7A. Carbachol reduced the peak amplitude of apamin-resistant tail currents by 25.8 ± 21% (from 128 ± 23 to 92 ± 33 pA; n = 6).

View larger version (18K): [in this window] [in a new window]   Fig. 7. Carbachol reduces the apamin-insensitive tail currents, reversibly inhibits mAHP and increases the firing rate of CA3 st. radiatum interneurons. A: superimposed tail currents recorded at the end of a 100-ms-long voltage step to +10 mV after apamin (100 nM) and after apamin + carbachol (40 µM) application. Note that carbachol furher reduces the apamin-insensitive current. B: action potential before (control), during carbachol (10 µM), and after atropine (1 µM) application. Carbachol reduces the mAHP without changing the action potential waveform. C: repetitive firing induced by prolonged depolarizing current pulses before, during carbachol application, and during co-application of carbachol and atropine. Carbachol increases the firing frequency. Atropine addition reverses the carbachol effect.

The effects of carbachol on action potential and firing rate of st. radiatum interneurons were studied in current-clamp mode. In five of six cells, carbachol (10 µM) depolarized the cell membrane by 4.4 ± 1.5 mV, while in one cell it did not affect the membrane potential. Hyperpolarizing current was therefore applied to restore the membrane potential to the control level, and, under these conditions, carbachol did not significantly change the action potential features but clearly reduced the mAHP amplitude (45 ± 22%) and duration (27 ± 19%, P = 0.028, n = 5; Fig. 7B). The effect was suppressed by atropine (1 µM), indicating that the response was mediated by muscarinic receptors (Fig. 7B). In three of five cells the suppression of mAHP induced an increase in firing rate (16 ± 4% change; n = 3, P = 0.036), an effect attenuated by atropine (Fig. 7C).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

To date, there is very little detailed characterization of the morphological and functional properties of the hippocampal interneurons located in st. radiatum. Moreover, very few studies have been performed on nonpyramidal neurons in the CA3 hippocampal subfield (Arancio et al. 1994; Gulyás et al. 1993; McBain and Dingledine 1993; Miles et al. 1996; Poncer et al. 1995). In the present study, we used intracellular staining, in situ hybridization, and whole cell recordings to provide a first detailed description of the intrinsic membrane properties and firing behavior of CA3 st. radiatum interneurons and outlined the role and the molecular nature of potassium conductances underlying the mAHP and regulating their discharge pattern.

Morphological features of CA3 st. radiatum interneurons

As observed by others (Gulyás et al. 1993; McMahon and Kauer 1997; Woodson et al. 1989), CA3 interneurons represent a heterogeneous population of cells with variable patterns of dendritic arborization, mostly confined to st. radiatum. As for their functional connectivity, CA3 st. radiatum interneurons are supposed to be contacted by afferents to pyramidal cell apical dendrites such as the commissural-associational fibers, which might provide a feed-forward drive (Gulyás et al. 1993), while input from recurrent collaterals of pyramidal cells to st. radiatum interneurons would mediate feedback inhibition (Freund and Buzsáki 1996).

In our study, in 3 of 15 slices from neonatal rats, labeled interneurons appeared to be dye-coupled, suggesting the presence of gap junctions. Gap junctions (Fukuda and Kosaka 2000; Katsumaru et al. 1988; Kosaka and Hama 1985) and dye-coupling (Michelson and Wong 1994; Strata et al. 1997) have been previously described in hippocampal interneurons.

Physiological properties of st. radiatum interneurons

Despite their heterogeneous morphological features, CA3 st. radiatum interneurons studied in the postnatal period P10-21 exhibited similar basic membrane properties, compatible with a complete postnatal development at P10. This is in line with the fact that interneurons appear early in the hippocampal formation, many of them before the principal cells (Altman and Das 1965; Amaral and Kurz 1985; Bayer 1980), in which developmental changes, mainly related to cell excitability and action potential properties (Costa et al. 1991; Spigelman et al. 1992), still occur up to P20.

Membrane properties of st. radiatum interneurons were found unambiguously different from those of pyramidal cells. In particular, a higher input resistance and shorter membrane time constant characterized these cells, in accordance with the data obtained from other hippocampal regions (Lacaille et al. 1987; Morin et al. 1996). Higher input resistances and shorter membrane time constant should be responsible for the generation of larger and faster synaptic potentials. Indeed, excitatory postsynaptic potentials (EPSPs) recorded from guinea pig st. radiatum interneurons were larger and faster than EPSPs of CA3 pyramidal cells (Miles 1990). Time constant and input resistance values found in the present study are higher than those reported for other interneurons (Morin et al. 1996; Williams et al. 1994), most likely due to the different recording conditions (patch-clamp vs. intracellular recording), temperature, and animal age. Anyway, it was reported that interneurons located in CA3 area have larger time constant and input resistance values than those reported in other hippocampal regions (Chitwood et al. 1999).

Spontaneous firing was found in st. oriens neurons (Maccaferri and McBain 1996; McBain 1994) but was absent in st. L-M interneurons (Lacaille and Scwartzkroin 1988) and axo-axonic cells (Buhl et al. 1994), as well as in the interneurons characterized in the present study.

In CA3 st. radiatum interneurons the maximum firing rate at the steady state was 31 Hz, comparable with that observed in neocortical "regular firing" nonpyramidal neurons (Erisir et al. 1999). The firing rate slightly decelerated over the first few spikes only. Little accommodation has also been found in CA1 st. L-M interneurons, which showed a voltage-dependent mode of firing (Lacaille and Schwartzkroin 1988), never observed in st. radiatum interneurons.

Inward rectification with anodal brake excitation was observed in CA3 st. radiatum interneurons and was abolished by external Cs⁺. In this respect, these cells resemble more closely nonpyramidal cells in st. oriens, in which both effects are mediated by the hyperpolarization-activated current I_h (Lacaille and Williams 1990; Maccaferri and McBain 1996).

Potassium conductances underlying the mAHP

Several types of K⁺ currents underlie the AHP in nerve cells. A Ca²⁺- and voltage-dependent, TEA-sensitive potassium current (I_C) (Adams et al. 1982) contributes to action potential repolarization and the fast and medium AHP (Storm 1987, 1989). At least two other potassium conductances contribute to the mAHP: I_AHP, due to the activation of small-conductance, Ca²⁺-activated, and voltage-insensitive SK channels of the apamin-sensitive type (Aoki and Baraban 2000; Sah 1996; Stocker et al. 1999); and I_M, a voltage-gated K⁺ current suppressed by muscarinic agonists (Halliwell and Adams 1982; Storm 1989). The slow AHP is instead mediated by sI_AHP, (Lancaster and Adams 1986; Sah 1996), an apamin-insensitive Ca²⁺-activated K⁺ current lasting more than 1 s and sensitive to modulation by many neurotransmitters.

As observed in many nonpyramidal cells (Aoki and Baraban 2000; for a review see Freund and Buzsáki 1996), CA3 st. radiatum interneurons fired action potentials followed by a medium-duration (~200 ms) mAHP, which may account for the slight accommodation observed in response to sustained current injection. Tail potassium currents underlying the mAHP were characterized in CA3 st. radiatum interneurons and did not show substantial differences in kinetics from those observed in pyramidal cells (Stocker et al. 1999), st. L-M interneurons (Aoki and Baraban 2000), as well as in hypoglossal motoneurons (Lape and Nistri 2000). As observed in other cells (see, for example, Hoshi and Aldrich 1988; Johansson et al. 1996), the reversal potential (E_rev) for the tail currents slightly deviated from the predicted value for K⁺. Anyway, a positive shift in E_rev was observed by raising the external K⁺ concentration. This suggested that the tail currents were mainly mediated by an increased membrane permeability to K⁺. The systematic difference between the measured E_rev and the predicted E_K could be due to incomplete equilibration of the pipette K⁺ with all cytosolic compartments, accumulation of K⁺ near the membrane, or interference by inward currents of other origin.

Apamin-sensitive potassium conductances

Apamin-sensitive potassium currents have been detected in the hippocampus in st. oriens (Zhang and McBain 1995) and in st. L-M (Aoki and Baraban 2000) interneurons as well as in CA1 pyramidal cells (Stocker et al. 1999).

Our results reveal that the mAHP of CA3 st. radiatum interneurons, as that observed in st. L-M interneurons of CA1 region, is mostly due to the activation of apamin-sensitive Ca²⁺-activated K⁺ channels, and that its reduction increases their firing discharge. The observed variability in the effect of apamin (range in different cells: from 17 to 79% reduction of I_AHP, n = 15) might suggest a certain variability in the level of expression of apamin-sensitive channels in subsets of CA3 str. radiatum interneurons. Nevertheless, the presence in most cells of a significant apamin-sensitive current component is supported by a high expression of mRNA coding for the highly apamin-sensitive SK2 subunit, as detected by in situ hybridization in CA3 str. radiatum nonpyramidal cells. As a note of caution, from the in situ experiments we cannot exclude that some of the labeled cells in the CA3 st. radiatum might correspond to cell types that were not characterized electrophysiologically in this study. The apamin-sensitive I_AHP does not contribute to action potential repolarization (Lape and Nistri 2000; Sah and McLachlan 1992; Stocker et al. 1999) but is activated during the mAHP following single action potentials. The apamin-sensitive tail current was characterized by a monoexponential deactivation with a time constant (~50 ms) in the same range as found in CA1 st. L-M hippocampal interneurons (~70 ms) (Aoki and Baraban 2000).

Carbachol-sensitive potassium conductances

Cholinergic innervation from the medial septum (Freund and Buzsáki 1996; Miettinen and Freund 1992) synaptically contacts hippocampal pyramidal cells and inhibitory neurons that contribute to theta rhythm generation by rhythmically inhibiting pyramidal cells (Tóth et al. 1997). Agonists of muscarinic receptors and stimulation of cholinergic afferents were found to increase the rate of occurrence of spontaneous inhibitory postsynaptic potentials (IPSPs) recorded in hippocampal pyramidal cells (Pitler and Alger 1992), suggesting their positive modulation of interneuron excitability. In all hippocampal regions, muscarine mostly excited interneurons (McQuiston and Madison 1999a; Parra et al. 1998), for example, by suppressing the AHP and enhancing the ADP (McQuiston and Madison 1999b).

In CA1 st. oriens (Zhang and McBain 1995), muscarinic agonists reduced the mAHP amplitude, increasing the firing discharge rate. It is interesting to note that, in st. oriens interneurons, carbachol did not alter the AHP in the presence of Ca²⁺-free solution (Zhang and McBain 1995). Our data reveal that CA3 st. radiatum interneurons possess functionally active muscarinic receptors that, on activation, induce a reduction of mAHP with a consequent increase in firing discharge. Apamin-insensitive tail potassium currents were reduced by carbachol, revealing that more than one type of potassium current contributes to the mAHP generation. A possible candidate to mediate the apamin-insensitive, muscarine-sensitive mAHP component is I_M, which has been shown to contribute to the mAHP for example in CA1 pyramidal neurons (Storm 1989).

In CA1 st. L-M interneurons, outward tail currents are dominated by a Ca²⁺-dependent current, blocked by apamin and TEA, largely insensitive to carbachol (Aoki and Baraban 2000). Different expression of potassium channels with distinct pharmacological sensitivity seems therefore to characterize the interneurons in various hippocampal fields, contributing to determine different firing pattern and conferring them subtype-specific properties.

In conclusion, our data show that st. radiatum interneurons of the CA3 hippocampal region represent a morphologically heterogeneous class of cells with similar membrane properties. High-input resistance, short lasting action potentials, and AHPs of medium duration discriminate them from pyramidal cells. Sustained firing allows them to be classified as "regular firing" cells, in which the apamin- and carbachol-sensitive potassium conductances underlying the mAHP are mainly shaping the firing pattern. Neurotransmitters and neuromodulators affecting mAHP can therefore differentially regulate the firing properties of these interneurons, mediating a control of hippocampal circuits.