Action of tachykinins in the hippocampus: Facilitation of inhibitory drive to GABAergic interneurons

Neuroscience 156 (2008) 527–536

ACTION OF TACHYKININS IN THE HIPPOCAMPUS: FACILITATION OF INHIBITORY DRIVE TO GABAergic INTERNEURONS R. OGIER,b1 L. J. WROBELa1 AND M. RAGGENBASSa*

numerous pyramidal neurons and can thus regulate the activity of a large neuronal population (Freund and Buzsaki, 1996). In contrast to pyramidal neurons, which constitute a relatively uniform population, interneurons appear to be highly heterogeneous, both in their morphology and in their physiological and pharmacological properties (Maccaferri and Lacaille, 2003; Mott and Dingledine, 2003; Monyer and Markram, 2004; Ascoli et al., 2008). In addition to control principal cells, hippocampal interneurons can also contact other interneurons (Freund and Buzsaki, 1996). Thus, calretinin-immunoreactive interneurons establish frequent connections with each other as well as with other subclasses of interneurons (Gulyas et al., 1996). Distinct populations of vasoactive intestinal polypeptide-containing interneurons display target selectivity for other interneurons (Acsady et al., 1996). GABAergic neurons projecting to the medial septum, and which are in part somatostatin-immunoreactive, also innervate interneurons in the CA1 and CA3 regions (Gulyas et al., 2003). Inhibitory interactions between hippocampal GABAergic interneurons and pyramidal neurons have been extensively investigated (Freund and Buzsaki, 1996). Less is known concerning inhibitory interactions between interneurons. Hippocampal interneurons possess GABAA receptors probably formed by a variety of combinations of ␣, ␤ and ␥ subunits (Gao and Fritschy, 1994; Patenaude et al., 2001). GABAergic synapses of interneurons display long-term activity-dependent plasticity distinct from that of pyramidal neurons (Patenaude et al., 2005). The inhibitory input to interneurons can be facilitated by glutamate, an effect mediated by presynaptic kainate receptors (Cossart et al., 2001). By contrast, activation of presynaptic metabotropic glutamate receptors (group III) can depress inhibitory postsynaptic currents (IPSCs) (Kogo et al., 2004). Functionally, inhibitory interactions between subclasses of interneurons may contribute to the generation of hippocampal theta and gamma rhythms (Whittington et al., 1995; Banks et al., 2000; White et al., 2000; Bartos et al., 2002). The hippocampus is innervated by substance P-containing axon terminals (Borhegyi and Leranth, 1997) and hippocampal GABAergic interneurons possess tachykinin receptors (Acsady et al., 1997; Sloviter et al., 2001). Peptides of the tachykinin family can powerfully excite hippocampal interneurons (Dreifuss and Raggenbass, 1986; Zaninetti and Raggenbass, 2000), an action which is mediated by neurokinin 1 (NK1) receptors and is due to a peptide-induced suppression of a resting K⫹ conductance (Ogier and Raggenbass, 2003). By exciting hippocampal interneurons, tachykinins enhance the frequency of spon-

a

Department of Basic Neurosciences, University Medical Center, 1, rue Michel-Servet, CH-1211 Geneva 4, Switzerland

b

Sanofi-Aventis (Suisse), CH-1217 Meyrin, Geneva, Switzerland

Abstract—By acting on neurokinin 1 (NK1) receptors, neuropeptides of the tachykinin family can powerfully excite rat hippocampal GABAergic interneurons located in the CA1 region and by this way indirectly inhibit CA1 pyramidal neurons. In addition to contact pyramidal neurons, however, GABAergic hippocampal interneurons can also innervate other interneurons. We thus asked whether activation of tachykinin-sensitive interneurons could indirectly inhibit other interneurons. The study was performed in hippocampal slices of young adult rats. Synaptic events were recorded using the whole-cell patch clamp technique. We found that substance P enhanced GABAergic inhibitory postsynaptic currents in a majority of the interneurons tested. Miniature, action potential–independent inhibitory postsynaptic currents were unaffected by substance P, as were evoked inhibitory synaptic currents. This suggests that the peptide acted at the somatodendritic membrane of interneurons, rather than at their axon terminals. The effect of substance P was mimicked by a selective NK1 receptor agonist, but not by neurokinin 2 (NK2) or neurokinin 3 (NK3) receptor agonists, and was suppressed by a NK1 selective receptor antagonist. In contrast to substance P, oxytocin, another peptide capable of activating hippocampal interneurons, had no effect on the inhibitory synaptic drive onto interneurons. We conclude that tachykinins, by acting on NK1 receptors, can influence the hippocampal activity by indirectly inhibiting both pyramidal neurons and GABAergic interneurons. Depending on the precise balance between these effects, tachykinins may either activate or depress hippocampal network activity. © 2008 IBRO. Published by Elsevier Ltd. All rights reserved. Key words: inhibitory postsynaptic current, oxytocin, pyramidal neurons, rat, substance P.

The CA1 region of the hippocampus contains two main classes of neurons: i) principal neurons, i.e. pyramidal neurons, which are glutamatergic and project to a variety of other brain regions; ii) interneurons, which are GABAergic and project locally. A single interneuron can synapse with 1

R.O. and L.J.W. contributed equally to this work. *Corresponding author. Tel: ⫹41-22-379-53-86; fax: ⫹41-22-379-54-02. E-mail address: [email protected] (M. Raggenbass). Abbreviations: IPSC, inhibitory postsynaptic current; mIPSC, miniature inhibitory postsynaptic current; MK801, (⫹)-5-methyl-10,11-dihydro5H-dibenzo[a,d]cyclohepten-5,10-imine maleate); NBQX, 6-nitro-7sulfamoylbenzo[f]quinoxaline-2,3-dione disodium; NK1, neurokinin 1; NK2, neurokinin 2; NK3, neurokinin 3; PBS, phosphate buffer saline; senktide, succinyl-(asp6,N-me-phe8)-substance P(6 –11); TTX, tetrodotoxin; ␤-ala-NKA, (␤-ala8)-neurokinin A(4 –10). 0306-4522/08 © 2008 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2008.08.001

527

528

R. Ogier et al. / Neuroscience 156 (2008) 527–536

taneous inhibitory events in principal cells, causing an indirect inhibition of these neurons (Ogier and Raggenbass, 2003). Due to the existence of interneuron–interneuron synapses, however, tachykinins may also indirectly inhibit other interneurons, leading to a partial disinhibition of principal cells. The aim of the present work was to determine whether tachykinins can synaptically inhibit hippocampal CA1 interneurons. Recordings were performed in acute hippocampal slices of young adult rats using the whole-cell patch clamp technique.

and decay time was measured from the peak by fitting with a single exponential. For inhibitory synaptic currents (IPSCs), interevent intervals were distributed exponentially or approximately exponentially. Their average and spread are expressed as mean and 95% confidence interval. The amplitudes of IPSCs were distributed lognormally (Zaninetti and Raggenbass, 2000; Ogier and Raggenbass, 2003). The mean ␮ and the standard deviation ␴ of amplitudes were computed using the following equations:

EXPERIMENTAL PROCEDURES

where M and S are the mean and the standard deviation of the normally distributed log-transformed amplitudes. For normally or lognormally distributed samples, average value and spread are given as mean⫾S.E.M. The significance of the effect on the spontaneous IPSC intervals and amplitudes was evaluated using the two-sample Kolmogorov-Smirnov test, whereas the significance on the IPSC kinetic parameters was evaluated using the Wilcoxon rank-sum test.

Hippocampal slices Experimental protocols conformed to the rules of the Swiss Veterinary Office and international guidelines on the ethical use of animals. Care was taken to minimize the number of animals used and their suffering. The animals used were 2 to 3-week-old rats of the Sprague–Dawley strain (Charles River Laboratories, Iffa Credo, L’Arbresle, France). They were anesthetized (pentobarbital i.p., 50 mg/kg) and killed by decapitation. The brain was removed and a block of tissue containing the dorsal part of the hippocampus was prepared. Coronal slices, 300 – 400-␮m-thick, were cut using a vibrating microtome (Campden Instruments, Leicester, UK). A slice was transferred to thermoregulated (32– 33 °C) recording chamber and perfused with a solution containing (in mM): NaCl, 135; NaHCO3, 15; KCl, 5; MgCl2, 1; CaCl2, 2; glucose 10; and saturated with 95% O2/5% CO2 (pH 7.3–7.4). Neurons were visualized using a Nikon Eclipse E600FN, equipped with 40⫻0.8 NA water-immersion objective and IR-DIC optics, and an infrared-sensitive videocamera (C 25400-07, Hamamatsu, Schüpfen, Switzerland). Unless otherwise stated, in order to isolate GABAergic IPSCs, the perfusion solution was supplemented with blockers of glutamatergic synaptic transmission, 6-nitro-7sulfamoylbenzo[f]quinoxaline-2,3-dione disodium (NBQX; 5 ␮M) and MK801 (20 ␮M).

Electrophysiological recordings Whole-cell recordings were performed in neurons located in the CA1 region using patch pipettes pulled from borosilicate glass capillaries (Harvard Apparatus, Les Ulis, France). To monitor synaptic activity, in either pyramidal neurons or interneurons, patch pipettes were filled with the following solution (mM): CsCl, 150; Hepes, 10; MgCl2, 4; 1,2,-bis(2-aminophenoxy)ethaneN,N,N=,N=-tetraacetic acid (BAPTA), 0.1; Na2-ATP, 2; Na2-GTP, 0.4 (pH 7.2–7.3, adjusted with NaOH). Some recordings were performed using a pipette solution in which CsCl was substituted with 140 mM K-gluconate and 10 mM KCl. The calculated liquid junction potentials were, respectively, 13 mV (negative) and 14.9 mV (positive; Clampex Junction Potential Calculator, pClamp software; Molecular Devices Sunnyvale, CA, USA). In some experiments, the pipette solution was supplemented with 1 mM QX-314 and/or 5 mM biocytin. The pipette resistance was 3–5 M⍀. Current signals, recorded in voltage clamp, were amplified, digitized and acquired using an Axopatch 200A amplifier, a Digidata 1200 interface and pClamp software (Molecular Devices). They were low-pass filtered online at 2 kHz and digitized at 10 kHz. In some experiments, synaptic events were evoked by delivering rectangular current pulses (100 ␮s, 0.2 Hz), using a two-pole twisted tungsten wire as a stimulation electrode. Spontaneous synaptic events were detected using the Mini Analysis package (Synaptosoft, Decatur, GA, USA). The threshold current was set to 10 –20 pA and the threshold charge transfer to 30 – 60 fC. The detected events were visually inspected and contaminating noise events were rejected. Rise time was 10 –90%

␮⫽exp共M⫹S2 ⁄ 2兲 and

␴2⫽exp共2M⫹2S2兲⫺exp共2M⫹S2兲

Biocytin labeling Slices containing biocytin-labeled neurons were incubated overnight in a solution of 4% paraformaldehyde dissolved in 0.15 M phosphate buffer saline (PBS), pH 7.4. They were then rinsed in PBS, immersed in 30% sucrose dissolved in the same buffer and stored at 4 °C until use. They were frozen and cut into 40-␮m-thick sections in a cryostat microtome. Biocytin was revealed using Cy2-conjugated streptavidin (No. 016-220-084; Jackson Immuno Research, West Grove, PA, USA), diluted at 1:50 in PBS. Labeled neurons were visualized using appropriate fluorescence filters. They were photographed, and their soma size measured, using a Nikon camera (Model DS-5M, with control unit DSL1; Nikon Corporation, Tokyo, Japan). Adobe Illustrator software was used to assemble the photomicrographs and to add scale bars and lettering.

Chemical compounds Substance P, (sar9,met(O2)11)-substance P, (␤-ala8)-neurokinin A(4 –10) (␤-ala-NKA) and succinyl-(asp6,N-me-phe8)-substance P(6 –11) (senktide) were purchased from Bachem (Bubendorf, Switzerland; for reviews on the pharmacology of tachykinin receptors, see Regoli et al., 1988, 1994). The NK1 receptor antagonist was kindly donated by Dr. E. Weinling (Sanofi-Synthélabo Recherche, Chilly-Mazarin, France). The effects of these compounds were tested by adding them to the perfusion solution. NBQX, (⫹)MK 801 and bicuculline methochloride were from Tocris Cookson (Bristol, UK). QX-314 was from Alomone Laboratories (Jerusalem, Israel) and biocytin from Sigma (St. Louis, MO, USA).

RESULTS Effect of tachykinins on GABAergic inhibitory transmission The hippocampus contains various subclasses of interneurons which innervate other interneurons (see introduction). We asked whether tachykinins, by exciting interneurons of this type, could cause an indirect inhibition of at least a subpopulation of interneurons. Whole-cell recordings were performed in CA1 interneurons located in radiatum layer. Under IR-DIC microscopy, multipolar or bipolar cells were visually identified. Large neurons having pyramidal shape were avoided, since they could represent non-inhibitory giant cells (Gulyas et al., 1998).

R. Ogier et al. / Neuroscience 156 (2008) 527–536

Fig. 1. Effect of substance P and bicuculline on spontaneous IPSCs in a hippocampal CA1 radiatum layer neuron. The current traces were obtained in voltage clamp in control conditions, in the presence of substance P (SP; 1 ␮M), in the presence of bicuculline (25 ␮M) and substance P, and following washout of both bicuculline and substance P. Note that bicuculline, a competitive GABAA receptor antagonist, caused a reversible suppression of inhibitory synaptic events.

CsCl-containing patch pipettes were used. Cells were held at ⫺60 to ⫺50 mV. Under these recording conditions, the average cell input resistance was 276⫾30 M⍀ (n⫽39) and IPSCs appeared as inward-going currents. Note that due to the presence of Cs⫹, a K⫹ channel blocker, the substance P-induced inward current was not detectable. Substance P, added to the perfusion solution at concentrations ranging from 20 nM to 1 ␮M, increased the frequency, and in most cases also the amplitude, of spontaneous IPSCs in a majority of the neurons tested (37 of 47 neurons; Figs. 1, 2A). When present, this effect was reversible and reproducible and tachyphylaxis was not observed. The recorded IPSCs were GABAergic. Indeed, in five of five neurons, the competitive GABAA receptor antagonist bicuculline (25 ␮M) suppressed the synaptic events recorded either in control conditions or in the presence of substance P (Fig. 1). The effect of substance P on IPSCs could be reliably quantified in 22 neurons, challenged with three different peptide concentrations, 1 ␮M, 0.1 ␮M and 20 nM (Fig. 2). In all cases, the peptide effect on the interevent interval was highly significant (P⬍0.001). In 14 neurons, in which substance P was applied at 1 ␮M, the mean IPSC interval was 156⫾25 ms in control conditions (range, 54 –306 ms), whereas in the presence of the peptide it decreased to 63⫾15 ms (range, 31–168 ms). In four neurons, substance P was tested at 0.1 ␮M. The control IPSC interval was 112⫾27 ms (range, 58 –179 ms), whereas in the presence of the peptide the IPSC interval was 59⫾17 ms (range, 28 –99 ms). Four neurons were challenged with 20 nM

529

substance P. The IPSC interval was 255⫾75 ms in control conditions (range, 62–394 ms) and 145⫾45 ms in the presence of the peptide (range, 38 –230 ms). Substance P also caused a significant increase in the IPSC amplitude in a majority of these neurons (P⬍0.001 in each case; Figs. 2 and 3B). In 10 of 14 neurons challenged with 1 ␮M substance P, the IPSC amplitude increased from 31⫾4 pA (range, 15–51 pA) to 59⫾8 pA (range, 21– 87 pA). In three of four neurons tested with 0.1 ␮M substance P, the increase in IPSC amplitude was from 38⫾14 pA (range, 19 – 65 pA) to 58⫾15 pA (range, 27–75 pA). In all four neurons challenge with 20 nM substance P, the IPSC amplitude increased from 30⫾7 pA (range, 10 – 41 pA) to 43⫾9 pA (range, 14 –59 pA). The charge carried by the substance P–induced synaptic inhibition (1 ␮M) was about three times larger than that transferred during basal inhibitory activity. If expressed in term of average inhibitory current, it was 12⫾4 pA in control conditions and 34⫾4 pA in the presence of the peptide (n⫽5; measured over a 2-s interval; Fig. 3A). The rise time of spontaneous IPSCs ranged from 1.04 –1.97 ms and was, on average, 1.4⫾0.1 ms. Substance P did not alter the rise time in seven neurons, decreased it slightly in three neurons (17%, 10% and 26%; P⬍0.01) and increased it slightly in two neurons (19% and 17%; P⬍0.01). The IPSC decay time ranged from 1.5– 4.4 ms, with an average value of 2.4⫾0.3 ms. It was unaffected by substance P in three neurons, was increased in eight neurons (8 –77%; P⬍0.01) and was decreased in one neuron (35%, P⬍0.01). Miniature, action potential-independent inhibitory postsynaptic currents (mIPSCs) were recorded in CA1 radiatum layer neurons in the presence of tetrodotoxin (TTX; 1 ␮M). Four neurons were examined. The mean mIPSC interval was 213⫾64 ms (range, 80 –389 ms) and the mean mIPSC amplitude was 31⫾2 pA (range, 26 –35 pA; Fig. 4). Substance P (1 ␮M) did not affect either the interval or the amplitude of mIPSCs in any of these neurons. The mIPSC rise time ranged from 1.4 –1.6 ms, and was 1.5⫾0.1 ms on average. The decay time was 2.6 –3.2 ms and was 2.7⫾0.2 ms on average. Neither the rise time nor the decay time was affected by substance P in any of these neurons. The decay time of mIPSCs was much faster than that of mIPSCs in pyramidal neurons, which was 10⫾1.2 ms (Ogier and Raggenbass, 2003). Similar data have also been obtained by others (Hajos and Mody, 1997; Nusser et al., 2001). Pharmacology of the tachykinin action The pharmacological profile of the receptors responsible for the facilitatory effect of substance P was determined by making use of selective agonists and an antagonist of tachykinin receptors. Neither ␤-ala-NKA (0.1–1 ␮M; n⫽5) nor senktide (0.1–1 ␮M; n⫽5), agonists of neurokinin 2 (NK2) and neurokinin 3 (NK3) receptors respectively, could mimic the effect of substance P on the IPSC frequency or amplitude (Fig. 5). By contrast, the facilitatory effect of substance P on IPSCs could be mimicked by (sar9,met(O2)11)-substance P (0.1–1 ␮M; n⫽4), a

530

R. Ogier et al. / Neuroscience 156 (2008) 527–536

Fig. 2. Effect of substance P on the frequency and amplitude of spontaneous IPSCs in a hippocampal CA1 radiatum layer neuron. (A) Each panel shows four consecutive current traces recorded in the voltage-clamp configuration in the normal perfusion solution (control), in the presence of substance P (SP; 1 ␮M) and following washout of the peptide (recovery). (B, top panel) Cumulative plot of the IPSC interval distribution in control conditions (continuous line; n⫽195) and in the presence of substance P (dotted line; n⫽320). (B, bottom panel) Cumulative plot of the IPSC amplitude distribution in control conditions (continuous line) and in the presence of substance P (dotted line). Note that in the presence of the peptide, the IPSC interval decreased, whereas the IPSC amplitude increased.

NK1 receptor agonist, and could be suppressed by SR 140333 (1 ␮M; n⫽2), an NK1 receptor antagonist (Fig. 6). Thus, substance P enhanced inhibitory transmission in CA1 radiatum layer neurons by acting on NK1 receptors. Effect of tachykinins on evoked IPSCs Neurons were recorded with patch pipettes filled with a CsCl-containing solution (holding potential ⫺70/⫺60 mV). The stimulation electrode was placed in radiatum layer, a few hundred ␮m away from the recorded neuron. The intensity of the electrical stimulation was adjusted so that no failures occurred. Since glutamatergic synaptic transmission was blocked and since the stimulation was local, the evoked IPSC was probably monosynaptic (Thompson, 1994). In four out of four neurons, the amplitude of the evoked IPSC was 465⫾123 pA (range, 150 –737 pA), the time to peak was 2.6⫾0.6 ms (range, 1.3–3.7 ms) and the decay time 10.2⫾1.7 ms (range, 7.1–14.4 ms; Fig. 7). None of these parameters was significantly affected by substance P (0.5 ␮M).

Effect of oxytocin on spontaneous IPSCs It has been previously shown that oxytocin receptor agonists can stimulate inhibitory synaptic transmission in CA1 pyramidal neurons by activating interneurons located in stratum pyramidale (Zaninetti and Raggenbass, 2000). We tested whether these compounds also facilitated IPSCs in CA1 radiatum layer neurons. We found that oxytocin, added to the perfusion solution at concentrations ranging from 0.5 to 1 ␮M, did not alter the frequency or the amplitude of IPSCs in any of eight neurons tested. Morphological identification of CA1 interneurons Twenty CA1 radiatum layer neurons, recorded with patch pipettes filled with a biocytin-containing solution, were successfully labeled. In 16 of them, substance P (20 nM to 1 ␮M) induced an increase in the spontaneous IPSC frequency (see above). The remaining four were not challenged with the peptide or were unaffected by it. Histological examination confirmed that the labeled neurons were all located in CA1 radiatum layer. Most of them were positioned in its middle part or in the vicinity of stratum

R. Ogier et al. / Neuroscience 156 (2008) 527–536

Fig. 3. Effect of substance P on the average inhibitory current and on IPSC amplitude in a hippocampal CA1 radiatum layer neuron. (A) The average inhibitory current was computed, in control conditions (top panel) and in the presence of substance P (SP; 1 ␮M; bottom panel), by subtracting the baseline current (top line) from the average current generated by spontaneous IPSCs (bottom line). Note that substance P caused a threefold increase in the average inhibitory current, from 11 to 30 pA. (B) Histogram plot of the IPSC amplitude in control conditions (filled bars) and in the presence of the peptide (open bars). Note the substance P–induced increase in the amplitude of spontaneous inhibitory synaptic events.

lacunosum-moleculare (n⫽16; Fig. 8), whereas some were next to stratum pyramidale (n⫽4). Most of the neurons were multipolar (n⫽16), some were bipolar (n⫽2). In two instances, only a single large dendrite was visible. Their cell bodies had diameters ranging from 6 to 16 ␮m, with an average value of 12⫾1 ␮m. These data suggest that most, if not all, of the recorded neurons were CA1 GABAergic interneurons (Freund and Buzsaki, 1996). Effect of tachykinins on CA1 pyramidal neurons In our previous study, we found that low concentrations of substance P (10 –50 nM) could indirectly inhibit CA1 pyramidal neurons by enhancing GABAA-mediated synaptic

531

transmission (Ogier and Raggenbass, 2003). We asked whether high concentrations of tachykinins (0.5–1 ␮M) could induce an opposite effect, i.e. cause disinhibition of principal cells. Indeed, one could conjecture that if the tachykinin concentration is high enough, interneuron excitation could be so strong as to cause action potential inactivation and cessation of action potential-dependent GABA release. In a first series of experiments, four pyramidal neurons were recorded in the presence of blockers of glutamatergic synaptic transmission, using CsCl-containing patch pipettes. Substance P (0.5 ␮M) enhanced the inhibitory input to all four neurons (Fig. 9A). In the presence of the peptide, the IPSC interval decreased, from 99⫾21 ms (range, 51– 152 ms) to 36⫾9 ms (range, 20 –55 ms) and the IPSC amplitude increased from 43⫾7 pA (range, 28 –57 pA) to 91⫾15 pA (range, 56 –123 pA; P⬍0.05 in all cases). In a second series of experiments, glutamate receptor antagonists were omitted from the perfusion solution and pyramidal neurons were recorded with K-gluconate-containing pipettes. The holding potential was set to 0 mV. Under these conditions, both GABAergic and glutamatergic synaptic transmission was functional, excitatory postsynaptic currents were undetectable and recorded IPSCs appeared as outward currents. Five neurons were tested. Due to the recording conditions, the current trace was noisy and, as a consequence, the signal-to-noise ratio was less good than with CsCl-containing pipettes (see above). In control conditions, observable synaptic events were relatively rare (Fig. 9B, top trace), a fact that can probably in part be accounted by inefficient space clamp of dendrites. However, substance P at 1 ␮M induced an increase in the frequency of inhibitory synaptic events in all five neurons (Fig. 9B, bottom trace). The peptide effect could be reliably quantified in four neurons. Whereas in control conditions the IPSC interval was 1922⫾477 ms (range, 898 –3119 ms), it decreased to 520⫾256 ms in the presence of substance P (range, 125–1260 ms; P⬍0.05 in all cases). Taken together, these data indicate that substance P, at concentrations as high as 0.5–1 ␮M, was still able to enhance GABAergic inhibitory input to pyramidal neurons. These data do not support the notion that high concentrations of substance P may cause disinhibition of principal cells, and this irrespective of the presence or not of functional glutamatergic synaptic transmission within the hippocampal slice.

DISCUSSION Tachykinin inhibition of interneurons We have found that tachykinins can facilitate GABAergic IPSCs in the CA1 radiatum layer interneurons, presumably by exciting GABAergic interneurons that selectively innervate other interneurons (Freund and Buzsaki, 1996). Previous data from our laboratory showed that, by exciting GABAergic interneurons that innervate CA1 pyramidal neurons, tachykinins can enhance synaptic inhibition of the latter (Ogier and Raggenbass, 2003). Together, these results suggest that, depending on the specific interneuron

532

R. Ogier et al. / Neuroscience 156 (2008) 527–536

Fig. 4. Effect of substance P on the frequency and amplitude of spontaneous IPSCs and mIPSCs in a hippocampal CA1 radiatum layer neuron. (A) Current traces recorded in voltage-clamp in the normal perfusion solution (control), in the presence of substance P (SP; 1 ␮M), following addition of TTX (1 ␮M) to the perfusion solution and in the presence of both TTX and substance P. (B, top panels) Cumulative plots of spontaneous IPSC interval and amplitude distribution in control conditions (continuous line; n⫽65) and in the presence of substance P (dotted line; n⫽493). (B, bottom panels) Cumulative plots of mIPSC interval and amplitude distribution in control conditions (continuous line; n⫽150) and in the presence of substance P (dotted line; n⫽122). Note that substance P facilitated spontaneous IPSCs but not mIPSCs.

population(s) they activate, tachykinins can potentially influence hippocampal activity in a dual manner, by either depressing or disinhibiting principal cell activity. Substance P enhanced the frequency of spontaneous IPSCs, but did not affect the frequency of mIPSCs, recorded in the presence of TTX. In addition, substance P did not significantly modify the amplitude or kinetics of evoked IPSCs. The simplest explanation is that the peptide-facilitated IPSCs were action potential-dependent, i.e. were due to substance P acting on tachykinin receptors located on the somatodendritic membrane of interneurons, rather than on their axon terminals. Our data, however, do not exclude the presence of tachykinin receptors on interneuron axon terminals. Agonist binding to such receptors could have no effect per se (and would thus not affect mIPSCs), but could influence synaptic transmission in the presence of other concomitant events, such as presynaptic voltage-gated Ca2⫹ channel activation. In addition to increasing the frequency of spontaneous IPSCs in interneurons, substance P also increased their amplitude. This may suggest that substance P acted in part postsynaptically. However, the fact that the amplitude of either mIPSCs or evoked IPSCs was not modified in the presence of substance P argues against such a possibility. We suggest that the increase in spontaneous IPSCs amplitude was a consequence of the peptide-induced increase in the IPSC frequency. This may be due to IPSC summation, or to a frequency-dependent facilitation at GABAergic interneuron–interneuron synapses (Gupta et al., 2000), or to activation of previously quiescent presynaptic interneurons. This would also explain why, in some

instances, substance P induced a modification in the kinetic parameters of spontaneous IPSCs. The substance P–induced indirect inhibition of interneurons was mediated by NK1, but not NK2 or NK3 receptors. Thus, interneurons that innervate other interneurons possess the same type of tachykinin receptors as interneurons innervating pyramidal neurons (Ogier and Raggenbass, 2003). Our electrophysiological data are in accordance with morphological data showing that virtually all the hippocampal interneurons which are immunoreactive for parvalbumin, calbindin or calretinin, as well as for neuropeptides such as somatostatin, neuropeptide Y, cholecystokinin or vasoactive intestinal peptide, express tachykinin receptors of the NK1 type (Acsady et al., 1997; Sloviter et al., 2001). Tachykinins and epilepsy Tachykinins appear to play a critical role in the control of hippocampal excitability and seizures. Intrahippocampal administration of substance P, followed by perforant path stimulation, can induce and sustain a status epilepticus, and this effect can be blocked by a substance P antagonist (Liu et al., 1999b). In addition, mice lacking the preprotachykinin A gene, which encodes substance P, are resistant to kainate-induced seizures and excitotoxicity (Liu et al., 1999a). The seizure promoting effect of tachykinins may be explained as following. i) Due to possible differences in the intrinsic pharmacological/physiological properties of different types of interneurons, tachykinins could exert a more powerful effect on interneuron-specific inter-

R. Ogier et al. / Neuroscience 156 (2008) 527–536

533

dently of the presence or absence of glutamatergic synaptic transmission. One should point out, nevertheless, that in an in vitro hippocampal preparation part of the neuronal connections are severed. The impact of substance P on interneuron-mediated interneuron inhibition is thus probably limited. The effect of substance P may be drastically different in the in situ hippocampus, where the whole set of synaptic connections is functional. Oxytocin and interneurons The nonapetide oxytocin is synthesized in the supraoptic and paraventricular nuclei of the hypothalamus and can

Fig. 5. Effect of substance P and of NK2 and NK3 receptor agonists on the frequency and amplitude of spontaneous IPSCs in a hippocampal CA1 radiatum layer neuron. The cumulative plots of spontaneous IPSC interval (A) and amplitude (B) distribution were obtained in control conditions (continuous lines; n⫽120, 152 and 110, respectively) and in the presence of the following compounds (dotted lines): substance P (SP; 0.1 ␮M, n⫽283), ␤-ala-NKA (0.1 ␮M; n⫽138) and senktide (0.1 ␮M; n⫽118). Note that, contrary to substance P, the NK2 and NK3 receptor agonists did not affect the frequency or amplitude of spontaneous IPSCs.

neurons than on other interneurons. This would cause a preferential inhibition of interneurons controlling pyramidal neurons, thus leading to pyramidal neuron disinhibition. ii) Alternatively, since virtually all hippocampal interneurons possess NK1 receptors, tachykinins can excite the whole interneuronal population. If the tachykinin concentration is sufficiently high, interneuron excitation could be so strong as to cause action potential inactivation, cessation of GABA release and pyramidal neuron disinhibition. The data obtained in the present work do not support the conjecture just mentioned. We found that even at relatively high concentrations (of the micromolar order), substance P was still able to enhance the GABAergic inhibitory input to pyramidal neurons, and this indepen-

Fig. 6. Effect of a NK1 receptor agonist, of substance P and of a NK1 receptor antagonist on the frequency and amplitude of spontaneous IPSCs in a hippocampal CA1 radiatum layer neuron. The cumulative plots of spontaneous IPSC interval (A) and amplitude (B) distribution were obtained in control conditions (continuous lines, n⫽186, 174 and 180, respectively) and in the presence of the following compounds (dotted lines): (sar9,met(O2)11)-substance P (sar9-SP; 1 ␮M; n⫽311), substance P (SP; 1 ␮M; n⫽279) and SR 140333 and substance P (antag⫹SP; both at 1 ␮M; n⫽162). Note that the effect of substance P was mimicked by the NK1 receptor agonist and was suppressed by the NK1 receptor antagonist.

534

R. Ogier et al. / Neuroscience 156 (2008) 527–536

Fig. 7. Effect of substance P on the evoked IPSC in a hippocampal CA1 radiatum layer neuron. Each trace is the average of 20 individual responses. The recordings were performed in the normal perfusion solution (control; top panel) and in the presence of substance P (SP; 0.5 ␮M; middle panel). In the bottom panel, the control (solid line) and the test trace (dotted line) are superimposed. The open triangles mark the stimulus artifact. Note that substance P enhanced the spontaneous inhibitory activity, but did not modify significantly the amplitude or the kinetics of the evoked inhibitory current.

act both as a hormone and as a neurotransmitter/neuromodulator (Gimpl and Fahrenholz, 2001; Raggenbass, 2001). In a previous study, we found that oxytocin could cause an indirect inhibition of CA1 pyramidal neurons by exciting hippocampal CA1 interneurons located in stratum pyramidale but not in other strata (Zaninetti and Raggen-

Fig. 9. Effect of substance P on spontaneous IPSC in CA1 pyramidal neurons. (A) Current traces recorded in voltage-clamp in the normal perfusion solution (control; top trace) and in the presence of substance P (SP; 1 ␮M; bottom trace). The neuron was recorded using a CsClcontaining pipette solution and was held at ⫺70 mV. The perfusion solution was supplemented with NBQX (5 ␮M) and MK801 (20 ␮M). (B) Current traces recorded in voltage clamp in a second neuron, in the normal perfusion solution (control; top trace) and in the presence of substance P (SP; 1 ␮M; bottom trace). The recording was performed using a K-gluconate containing pipette solution. The neuron was held at 0 mV and blockers of glutamatergic synaptic transmission were omitted from the perfusion solution. Note that substance P enhanced spontaneous inhibitory synaptic activity in both neurons.

bass, 2000). This effect was mediated by uterine-type oxytocin receptors. In the present study we tested whether oxytocin, by acting on hippocampal interneurons, could also inhibit other interneurons. In contrast to what was found for tachykinins, oxytocin did not facilitate IPSCs in any of the recorded radiatum layer interneurons. This suggests that interneurons belonging to the IS-1 class, and which selectively innervate calbindin- or calretinin-contain-

Fig. 8. Biocytin-labeled substance P-responsive hippocampal CA1 neuron in a transverse hippocampal section. The photomicrographs show the labeled neuron at low (A) and high (B) magnification. In this neuron, substance P (1 ␮M) induced a decrease in the IPSC interval from 135⫾10 ms (n⫽247) to 66⫾2 ms (n⫽1367; P⬍0.001), whereas the IPSC amplitude (32⫾2 pA) was not significantly modified. Abbreviations: Py, pyramidal cell layer; Rad, radiatum layer; LMol, lacunosum moleculare layer; hif, hippocampal fissure. Scale bars⫽100 ␮m.

R. Ogier et al. / Neuroscience 156 (2008) 527–536

ing interneurons of radiatum layer (Freund and Buzsaki, 1996), probably possess NK1 tachykinin receptors but are devoid of oxytocin receptors.

CONCLUSION The present data as well as previous data from our and other laboratories indicate that hippocampal interneurons, in addition to glutamatergic, GABAergic, cholinergic and aminergic receptors (Freund and Buzsaki, 1996), possess receptors for neuropeptides such as tachykinins, oxytocin, cholecystokinin (CCK), thyrotropin-releasing hormone (TRH), somatostatin, vasoactive intestinal peptide (VIP) and opioid peptides (Raggenbass et al., 1985; Madison and Nicoll, 1988; Scharfman and Schwartzkroin, 1988; Atzori and Nistri, 1996; Miller et al., 1997; Tallent and Siggins, 1997; Yanovsky et al., 1997; Zaninetti and Raggenbass, 2000; Ogier and Raggenbass, 2003; Cunha-Reis et al., 2004; Deng et al., 2006). The former six peptides exert an excitatory action on interneurons, whereas opioid peptides have an inhibitory effect. Interneurons play a role in generating rhythmic activity in principal cells and can modify synaptic plasticity (Freund and Buzsaki, 1996; Whittington and Traub, 2003; Bartos et al., 2007; Mann and Paulsen, 2007). Thus, modulation of interneurons by neuropeptides may deeply influence the behavior of hippocampal networks. However, the effect of a neuropeptide on global hippocampal activity is difficult to predict, since it depends on a number of factors, among which: (i) the subpopulation of interneurons responding to the neuropeptide; (ii) the target neuronal population of the peptide responsive interneurons; (iii) the cellular location, strength and temporal dynamics of synapses between the responsive interneurons and their targets. For example, the effect of a given neuropeptide on the collective behavior of hippocampal principal cells will depend, among other things, on whether or not interneurons responsive to this neuropeptide make contact with other interneurons, via either GABAergic synapses or gap junctions (Tamas et al., 2000); or on whether interneurontarget cell synapses show short-term facilitation or depression (Gupta et al., 2000). In order to understand, even in a purely qualitative manner, the effect of neuropeptides on hippocampal activity, more detailed information concerning the intrinsic inhibitory circuitry of the hippocampus is needed. Acknowledgments—This work was supported in part by the Swiss National Science Foundation (grant to M.R.). R.O. was supported by the Swiss National Science Foundation MD-PhD program (Foundation Prof. Dr. Max Cloëtta, Zurich, Switzerland). We thank Dr. E. Weinling (Sanofi-Synthélabo Recherche, Chilly-Mazarin, France) for the supply of the NK1 receptor antagonist. We also thank Ms. D. Machard and A. Dupré for excellent technical assistance.

REFERENCES Acsady L, Gorcs TJ, Freund TF (1996) Different populations of vasoactive intestinal polypeptide-immunoreactive interneurons are specialized to control pyramidal cells or interneurons in the hippocampus. Neuroscience 73:317–334.

535

Acsady L, Katona I, Gulyas AI, Shigemoto R, Freund TF (1997) Immunostaining for substance P receptor labels GABAergic cells with distinct termination patterns in the hippocampus. J Comp Neurol 378:320 –336. Ascoli GA, Alonso-Nanclares L, Anderson SA, Barrionuevo G, Benavides-Piccione R, Burkhalter A, Buzsaki G, Cauli B, Defelipe J, Fairen A, Feldmeyer D, Fishell G, Fregnac Y, Freund TF, Gardner D, Gardner EP, Goldberg JH, Helmstaedter M, Hestrin S, Karube F, Kisvarday ZF, Lambolez B, Lewis DA, Marin O, Markram H, Munoz A, Packer A, Petersen CC, Rockland KS, Rossier J, Rudy B, Somogyi P, Staiger JF, Tamas G, Thomson AM, Toledo-Rodriguez M, Wang Y, West DC, Yuste R (2008) Petilla terminology: nomenclature of features of GABAergic interneurons of the cerebral cortex. Nat Rev Neurosci 9:557–568. Atzori M, Nistri A (1996) Effects of thyrotropin-releasing hormone on GABAergic synaptic transmission of the rat hippocampus. Eur J Neurosci 8:1299 –1305. Banks MI, White JA, Pearce RA (2000) Interactions between distinct GABA(A) circuits in hippocampus. Neuron 25:449 – 457. Bartos M, Vida I, Frotscher M, Meyer A, Monyer H, Geiger JR, Jonas P (2002) Fast synaptic inhibition promotes synchronized gamma oscillations in hippocampal interneuron networks. Proc Natl Acad Sci U S A 99:13222–13227. Bartos M, Vida I, Jonas P (2007) Synaptic mechanisms of synchronized gamma oscillations in inhibitory interneuron networks. Nat Rev Neurosci 8:45–56. Borhegyi Z, Leranth C (1997) Substance P innervation of the rat hippocampal formation. J Comp Neurol 384:41–58. Cossart R, Tyzio R, Dinocourt C, Esclapez M, Hirsch JC, Ben-Ari Y, Bernard C (2001) Presynaptic kainate receptors that enhance the release of GABA on CA1 hippocampal interneurons. Neuron 29:497–508. Cunha-Reis D, Sebastiao AM, Wirkner K, Illes P, Ribeiro JA (2004) VIP enhances both pre- and postsynaptic GABAergic transmission to hippocampal interneurones leading to increased excitatory synaptic transmission to CA1 pyramidal cells. Br J Pharmacol 143: 733–744. Deng PY, Porter JE, Shin HS, Lei S (2006) Thyrotropin-releasing hormone increases GABA release in rat hippocampus. J Physiol 577:497–511. Dreifuss JJ, Raggenbass M (1986) Tachykinins and bombesin excite non-pyramidal neurones in rat hippocampus. J Physiol 379:417–428. Freund TF, Buzsaki G (1996) Interneurons of the hippocampus. Hippocampus 6:347– 470. Gao B, Fritschy JM (1994) Selective allocation of GABAA receptors containing the alpha 1 subunit to neurochemically distinct subpopulations of rat hippocampal interneurons. Eur J Neurosci 6:837– 853. Gimpl G, Fahrenholz F (2001) The oxytocin receptor system: structure, function, and regulation. Physiol Rev 81:629 – 683. Gulyas AI, Hajos N, Freund TF (1996) Interneurons containing calretinin are specialized to control other interneurons in the rat hippocampus. J Neurosci 16:3397–3411. Gulyas AI, Hajos N, Katona I, Freund TF (2003) Interneurons are the local targets of hippocampal inhibitory cells which project to the medial septum. Eur J Neurosci 17:1861–1872. Gulyas AI, Toth K, McBain CJ, Freund TF (1998) Stratum radiatum giant cells: a type of principal cell in the rat hippocampus. Eur J Neurosci 10:3813–3822. Gupta A, Wang Y, Markram H (2000) Organizing principles for a diversity of GABAergic interneurons and synapses in the neocortex. Science 287:273–278. Hajos N, Mody I (1997) Synaptic communication among hippocampal interneurons: properties of spontaneous IPSCs in morphologically identified cells. J Neurosci 17:8427– 8442. Kogo N, Dalezios Y, Capogna M, Ferraguti F, Shigemoto R, Somogyi P (2004) Depression of GABAergic input to identified hippocampal

536

R. Ogier et al. / Neuroscience 156 (2008) 527–536

neurons by group III metabotropic glutamate receptors in the rat. Eur J Neurosci 19:2727–2740. Liu H, Cao Y, Basbaum AI, Mazarati AM, Sankar R, Wasterlain CG (1999a) Resistance to excitotoxin-induced seizures and neuronal death in mice lacking the preprotachykinin A gene. Proc Natl Acad Sci U S A 96:12096 –12101. Liu H, Mazarati AM, Katsumori H, Sankar R, Wasterlain CG (1999b) Substance P is expressed in hippocampal principal neurons during status epilepticus and plays a critical role in the maintenance of status epilepticus. Proc Natl Acad Sci U S A 96:5286 –5291. Maccaferri G, Lacaille JC (2003) Interneuron Diversity series: Hippocampal interneuron classifications-making things as simple as possible, not simpler. Trends Neurosci 26:564 –571. Madison DV, Nicoll RA (1988) Enkephalin hyperpolarizes interneurones in the rat hippocampus. J Physiol 398:123–130. Mann EO, Paulsen O (2007) Role of GABAergic inhibition in hippocampal network oscillations. Trends Neurosci 30:343–349. Miller KK, Hoffer A, Svoboda KR, Lupica CR (1997) Cholecystokinin increases GABA release by inhibiting a resting K⫹ conductance in hippocampal interneurons. J Neurosci 17:4994 –5003. Monyer H, Markram H (2004) Interneuron diversity series: Molecular and genetic tools to study GABAergic interneuron diversity and function. Trends Neurosci 27:90 –97. Mott DD, Dingledine R (2003) Interneuron diversity series: Interneuron research: challenges and strategies. Trends Neurosci 26:484–488. Nusser Z, Naylor D, Mody I (2001) Synapse-specific contribution of the variation of transmitter concentration to the decay of inhibitory postsynaptic currents. Biophys J 80:1251–1261. Ogier R, Raggenbass M (2003) Action of tachykinins in the rat hippocampus: modulation of inhibitory synaptic transmission. Eur J Neurosci 17:2639 –2647. Patenaude C, Massicotte G, Lacaille JC (2005) Cell-type specific GABA synaptic transmission and activity-dependent plasticity in rat hippocampal stratum radiatum interneurons. Eur J Neurosci 22:179 –188. Patenaude C, Nurse S, Lacaille JC (2001) Sensitivity of synaptic GABA(A) receptors to allosteric modulators in hippocampal oriensalveus interneurons. Synapse 41:29 –39. Raggenbass M (2001) Vasopressin- and oxytocin-induced activity in the central nervous system: electrophysiological studies using invitro systems. Prog Neurobiol 64:307–326. Raggenbass M, Wuarin JP, Gahwiler BH, Dreifuss JJ (1985) Opposing effects of oxytocin and of a mu-receptor agonistic opioid peptide on

the same class of non-pyramidal neurones in rat hippocampus. Brain Res 344:392–396. Regoli D, Boudon A, Fauchere JL (1994) Receptors and antagonists for substance P and related peptides. Pharmacol Rev 46:551–599. Regoli D, Drapeau G, Dion S, Couture R (1988) New selective agonists for neurokinin receptors: pharmacological tools for receptor characterization. Trends Pharmacol Sci 9:290 –295. Scharfman HE, Schwartzkroin PA (1988) Further studies of the effects of somatostatin and related peptides in area CA1 of rabbit hippocampus. Cell Mol Neurobiol 8:411– 429. Sloviter RS, Ali-Akbarian L, Horvath KD, Menkens KA (2001) Substance P receptor expression by inhibitory interneurons of the rat hippocampus: enhanced detection using improved immunocytochemical methods for the preservation and colocalization of GABA and other neuronal markers. J Comp Neurol 430:283–305. Tallent MK, Siggins GR (1997) Somatostatin depresses excitatory but not inhibitory neurotransmission in rat CA1 hippocampus. J Neurophysiol 78:3008 –3018. Tamas G, Buhl EH, Lorincz A, Somogyi P (2000) Proximally targeted GABAergic synapses and gap junctions synchronize cortical interneurons. Nat Neurosci 3:366 –371. Thompson SM (1994) Modulation of inhibitory synaptic transmission in the hippocampus. Prog Neurobiol 42:575– 609. White JA, Banks MI, Pearce RA, Kopell NJ (2000) Networks of interneurons with fast and slow gamma-aminobutyric acid type A (GABAA) kinetics provide substrate for mixed gamma-theta rhythm. Proc Natl Acad Sci U S A 97:8128 – 8133. Whittington MA, Traub RD (2003) Interneuron diversity series: Inhibitory interneurons and network oscillations in vitro. Trends Neurosci 26:676 – 682. Whittington MA, Traub RD, Jefferys JG (1995) Synchronized oscillations in interneuron networks driven by metabotropic glutamate receptor activation. Nature 373:612– 615. Yanovsky Y, Sergeeva OA, Freund TF, Haas HL (1997) Activation of interneurons at the stratum oriens/alveus border suppresses excitatory transmission to apical dendrites in the CA1 area of the mouse hippocampus. Neuroscience 77:87–96. Zaninetti M, Raggenbass M (2000) Oxytocin receptor agonists enhance inhibitory synaptic transmission in the rat hippocampus by activating interneurons in stratum pyramidale. Eur J Neurosci 12:3975–3984.

(Accepted 2 August 2008) (Available online 7 August 2008)