Downregulation of KCC2 following LTP contributes to EPSP–spike potentiation in rat hippocampus

BBRC Biochemical and Biophysical Research Communications 343 (2006) 1209–1215 www.elsevier.com/locate/ybbrc

Downregulation of KCC2 following LTP contributes to EPSP–spike potentiation in rat hippocampus q Wei Wang b

a,b

, Neng Gong

a,b

, Tian-Le Xu

a,b,*

a School of Life Sciences, University of Science and Technology of China, Hefei 230027, China Institute of Neuroscience, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai 200031, China

Received 4 March 2006 Available online 20 March 2006

Abstract GABAergic synaptic inhibition plays a critical role in regulating long-term potentiation (LTP) of glutamatergic synaptic transmission and circuit output. The K+–Cl cotransporter 2 (KCC2) is an important factor in determining inhibitory GABAergic synaptic strength besides the contribution of GABAA receptor. Although much knowledge has been gained regarding activity-dependent downregulation of KCC2 in many pathological conditions, the potential change and contribution of KCC2 in LTP expression is still unknown. In this study, we found that downregulation of KCC2 was accompanied with the occurrence of LTP but not that of long-term depression in hippocampal CA1 region. Meanwhile, KCC2 level in CA3/DG and adjacent cortex was stable in the process of LTP expression in Schaffer collateral synapses. Blockade of NMDA receptor with APV not only prevented LTP induction also abolished the reduction of KCC2. Furthermore, the inhibition of KCC2 function with furosemide directly induced EPSP–spike (E–S) potentiation, an important component of LTP in hippocampus. The present data suggest a novel mechanism that LTP formation is accompanied by the downregulation of KCC2, which is underlying GABAergic strength and most likely contributes to the E–S potentiation following LTP. Ó 2006 Elsevier Inc. All rights reserved. Keywords: KCC2; LTP; GABAergic inhibition; NMDA receptor; E–S potentiation

Long-term potentiation (LTP), the long-lasting enhancement of synaptic transmission induced by tetanic stimulation of afferents, has been used as a cellular model of learning and memory [1,2]. There are two components during LTP: increased excitatory postsynaptic potential (EPSP) and the EPSP–spike (E–S) potentiation that means the increased ability of the EPSP to generate a spike [3,4]. Previous studies have indicated that modification of GABAergic inhibition plays an important role in both components of LTP. For example, direct inhibition of GABAA receptor (GABAAR) facilitates the induction of LTP and increases circuit activity [5–7]. Furthermore, the protocols q

This study was supported by the National Natural Science Foundation of China (Nos. 30125015 and 30321002) and the National Basic Research Program of China (2006CB500803) to T.-L. Xu. * Corresponding author. Fax: +86 21 54921735. E-mail addresses: [email protected], [email protected] (T.-L. Xu). 0006-291X/$ - see front matter Ó 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2006.03.038

for induction of LTP can also induce long-term depression (LTD) of GABAergic inhibition, which significantly contributes to the facilitation of LTP induction and the subsequent E–S potentiation [4,8]. Generally, GABAergic synaptic inhibition is modulated by the following mechanisms: changes in the structure and amount of GABAAR, the release of GABA, or the firing rate of GABAergic neurons [9,10]. For example, GABAergic LTD can be mediated by calcineurin-dependent dephosphorylation of postsynaptic GABAAR [4] or cannabinoid receptor-dependent reduction of presynaptic GABA release [8]. In addition, recent studies have indicated that postsynaptic electrochemical gradient of Cl is also very important for the modification of GABAergic transmission in many physiological and pathological conditions [11–19]. In adult neurons, K+–Cl cotransporter 2 (KCC2) is the major Cl extrusion mechanism responsible for the low intracellular Cl concentration [20]. Activity-dependent

1210

W. Wang et al. / Biochemical and Biophysical Research Communications 343 (2006) 1209–1215

downregulation of KCC2 seems to be a common mechanism for the reduction of GABAergic inhibition in certain conditions [17], such as pain [14], ischemia [15], and epilepsy [16]. However, it’s still unknown whether the downregulation of KCC2 is also involved in LTP formation and contributes to the subsequent E–S potentiation. The present study explores the possible change and role of KCC2 in LTP by field potential recording and immunoblotting methods in rat hippocampus. Materials and methods Experimental animal. The care and use of animals for these experiments followed the guidelines and protocols approved by the Institutional Animal Care and Use Committee of Institute of Neuroscience, Shanghai Institute for Biological Sciences, the Chinese Academy of Sciences. Hippocampal slice preparation. Transverse hippocampal slices (400 lm thickness) were prepared from 17-day-old male Sprague–Dawley rats. Animals were killed by decapitation in accordance with institutional regulations. Slices were obtained using a vibratome (ATF-1000, Leica instruments Ltd., Wetzlar, Germany) in ice-cold artificial cerebrospinal fluid (ACSF) and then kept at room temperature for at least 1.5 h before transfering to the recording chamber. The same ACSF was also used in recording, which contained (in mM): 119 NaCl, 2.5 KCl, 2.5 CaCl2, 1.25 KH2PO4, 1.3 MgSO4, 26.2 NaHCO3, and 11 D-glucose, saturated with 95% O2/5% CO2 (pH 7.4). All drugs were from Sigma. Electrophysiology. Extracellular recordings were performed in the CA1 region of the hippocampus. The stimulating electrode (bipolar platinum–iridium electrode) was placed at the Schaffer collateral pathway in the stratum radiatum. The field EPSPs (fEPSPs) and population spike (PS) were recorded via a glass micropipette filled with ACSF (1– 3 MX) placed in stratum radiatum and in the CA1 somatic layer, respectively. Test pulses were delivered every 30 s. Signals were amplified by an Axonpatch-200B amplifier (Axon Instruments, Foster City, CA) and filtered at 5 kHz. Data were acquired and analyzed using Clampfit 9.0 software (Axon Instruments). In present study, the stimuli intensity was adjusted to evoke 50% of the maximum fEPSP. Thetaburst stimulation (TBS) (consisted of 12 bursts of four pulses at 100 Hz, delivered at an interburst interval of 200 ms) and 4-train protocol (consisted of four 1 s trains of stimuli, at 100 Hz, delivered 5 min apart) were used to induce LTP, respectively. Low frequency stimulation (LFS) was used to induce LTD and consisted of 900 pulses at 3 Hz. Control slices were continuously recorded without TBS or LFS. In experiments involving tetanized slices, the slices were removed for KCC2 measurement from the recording chamber 5 min after the last stimulus was delivered. In drug perfusion test, slices were treated during the whole process of experiment. To measure E–S coupling, the PS amplitude was calculated between the negative peak and a line drawn on the top of the two positive peaks. The fEPSP slope was calculated by measuring the slope of the rising phase of PS [4]. The E–S curves were fitted with a sigmoidal equation using Origin 7.5 software. The E50 value was defined as the value of fEPSP slope at which the PS amplitude was 50% of its maximal response. Western-blot analysis. After measurement of circuit plasticity and output, CA1 region was isolated from the hippocampal slices and prepared as previously described [21]. Meanwhile, CA3/DG region and adjacent cortex were also collected, respectively. Prepared samples were separated by 7.5% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) and then electrotransferred onto PVDF membrane. The membrane was probed with anti-KCC2 antibody (polyclonal, 1:400 dilution, Upstate, USA) or anti-GAPDH (monoclonal, 1:1000 dilution, Kangchen, China). HRP-anti-rabbit or anti-mouse conjugated secondary antibodies (1:2000, Chemicon, USA) were applied to combine corresponding primary antibody, respectively. Immunoreactive bands were visualized by enhanced chemiluminescence (ECL, Pierce). Densito-

metric analysis was conducted using Molecular Analysis software of BioRad. The same experiments (with multiple experiment conditions) were repeated at least four times. Statistical analysis. Comparisons of two groups were done by t test. Three or more group comparisons were conducted using one-way analysis of the variance (ANOVA) with Duncan’s post hoc tests. All data are expressed as the means ± SEM. Differences were considered significant at P < 0.05. N represents the number of experiments.

Results Downregulation of KCC2 was accompanied with LTP in CA1 region Theta-burst stimulation is a broadly accepted protocol for LTP induction because it is intended to mimic endogenous theta frequency recorded in the brains of exploring rats [1,22]. Furthermore, TBS is very effective in producing stable and robust LTP [23–27]. Indeed, as shown in Fig. 1A (b and c), TBS induced a robust LTP (154 ± 4% of baseline, P < 0.05, Fig. 1A) in hippocampal CA1 region. It’s well known that change of GABAergic synaptic strength plays a critical role in LTP expression [4–6,8]. However, the change and potential contribution of KCC2 was neglected in the process of LTP. In the present study, we observed that the total amount of KCC2 in CA1 region at 1 h after LTP induction significantly decreased (0.70 ± 0.09 of control, P < 0.05, n = 11, Fig. 1B), while KCC2 level remained unchanged in control groups during the whole process of recording (data not shown). Furthermore, we noticed that KCC2 level also reduced to a relative stable level in the process of LTP maintenance even 3 h after induction (data not shown). For the sake of clarity, point of 1 h was collected for the analysis in our following experiments. To further gain insights into the synaptic specificity of KCC2 downregulation in LTP expression, we next examined the change of KCC2 level in CA3/DG region and adjacent cortex in the same slice. As shown in Figs. 2A and B, the amount of KCC2 in these regions kept in a steady level (0.92 ± 0.05 of control in CA3/DG region, P > 0.05; 1.02 ± 0.10 of control in adjacent cortex, P > 0.05; n = 4) after TBS on Schaffer collateral synapses. In order to evaluate that reduction of KCC2 is a general mechanism underlying LTP, another protocol for LTP induction was applied. As shown in Figs. 3A and B, 4-train protocol not only induced a robust LTP (166 ± 8% of baseline, P < 0.05, Fig. 3A) also significantly decreased KCC2 level (0.67 ± 0.15, of control, P < 0.05, n = 4, Fig. 3B) in CA1 region. Together, these results suggest that KCC2 downregulation may be a specific phenomenon and mechanism underlying LTP expression. Downregulation of KCC2 during LTP was NMDA receptor-dependent It is well known that NMDA receptor (NMDAR) is a critical molecule leading to LTP in hippocampal CA1 region [2]. Bath application of the NMDAR

W. Wang et al. / Biochemical and Biophysical Research Communications 343 (2006) 1209–1215

1211

Fig. 1. Downregulation of KCC2 was accompanied with LTP. (A) Theta burst stimulation induced a robust LTP of fEPSP in hippocampal CA1 region. Representative recordings before (trace 1) and after (trace 2) TBS are shown. Each point represents the mean fEPSP slope ± SEM normalized to the baseline. (B) Representative Western blots obtained from control and LTP slices are shown above summary results. Data were obtained from densitometry and the level of KCC2 was normalized to that of GAPDH. The expression of total KCC2 protein in isolated rat hippocampal CA1 region showed a significant downregulation at 1 h after LTP induction. *P < 0.05.

KCC2 following TBS-induced LTP was NMDARdependent. Downregulation of KCC2 was not accompanied with LTD expression In order to assure whether the downregulation of KCC2 was specific for LTP, we further examined the change of KCC2 level after LTD induction. As shown in Fig. 5, low frequency stimulation (LFS) successfully induced LTD (82 ± 4% of baseline, P < 0.05, Fig. 5A) but did not alter the amount of KCC2 (0.91 ± 0.07 of control, P > 0.05, Fig. 5B) at 1 h after LTD induction. These results indicate that downregulation of KCC2 was specifically accompanied with the occurrence of TBS-induced LTP in hippocampal CA1 region. Inhibition of KCC2 caused E–S potentiation

Fig. 2. Stable level of KCC2 in CA3/DG and adjacent cortex. (A) Representative blots obtained from control and TBS-treated slices are shown above summary results. (B) The amount of KCC2 in CA3/DG and adjacent cortex region remained unchanged even after TBS treatment in Schaffer collateral synapses.

antagonist, APV (50 lM) completely blocked the induction of LTP (101 ± 2% of baseline, P > 0.05, Fig. 4A). Meanwhile, APV abolished the reduction of KCC2 level (0.89 ± 0.09 of control, P > 0.05, Fig. 4B). Thus, the downregulation of the amount of

To explore the possible role of the KCC2 downregulation during LTP, we examined the effect of furosemide, an inhibitor of KCC2, in hippocampal slices. We found that 100 lM furosemide significantly increased the amplitude of population spike (PS) and caused a leftward shift of the E–S curve (Fig. 6B), an indicator of increased network excitability. The E50 value (see Materials and methods) of the E–S curve was significantly reduced from 0.75 ± 0.03 to 0.61 ± 0.02 (P < 0.05, n = 5, Fig. 6C) by furosemide. In this concentration, furosemide was thought to be specific for blocking KCC2 [28]. Consistently, we also observed that application of furosemide made a

1212

W. Wang et al. / Biochemical and Biophysical Research Communications 343 (2006) 1209–1215

Fig. 3. Change of KCC2 level in 4-train protocol induced LTP. (A) The 4-train protocol induced a robust LTP of fEPSP in hippocampal CA1 region. Representative recordings before (trace 1) and after (trace 2) 4-train protocol were shown. Each point represents the mean fEPSP slope ± SEM normalized to the baseline. (B) Representative Western blots obtained from control and LTP slices are shown above summary results. The amount of KCC2 protein in isolated rat hippocampal CA1 region showed a significant reduction at 1 h after LTP induction. *P < 0.05.

Fig. 4. Downregulation of KCC2 during LTP was NMDA receptor-dependent. (A) Application of APV (50 lM) abolished the LTP induction. (B) The same concentration of APV prevented the downregulation of KCC2 at 1 h after TBS (n = 5).

significantly positive shift of the reversal potential of chloride and obvious increase in circuit activity (data not shown). However, in order to exclude the potential side-effects of furosemide, another KCC2 inhibitor, DIOA was applied in our experiments. Consistently, DIOA (50 lM) also induced a leftward shift of E–S curve (data not shown) as furosemide. Thus, it is likely that the downregulation of KCC2 might contribute to the E–S potentiation following LTP.

Discussion The main findings of the present study are: first, the downregulation of KCC2 was specifically accompanied with TBS-induced LTP formation. Second, the inhibition of KCC2 contributed to the E–S potentiation. To our knowledge, it is the first evidence that LTP-induction stimulation significantly reduced KCC2 level, suggesting an

W. Wang et al. / Biochemical and Biophysical Research Communications 343 (2006) 1209–1215

1213

Fig. 5. Downregulation of KCC2 was not accompanied with LTD. (A) Low frequency stimulation (LFS) induced a LTD of fEPSP in hippocampal CA1 region. Representative recordings before (trace 1) and after (trace 2) LFS were shown. (B) Western-blot analysis of total KCC2 protein in isolated rat hippocampal CA1 region showed no significant changes of KCC2 at 1 h after LTD induction (n = 5).

Fig. 6. Inhibition of KCC2 caused EPSP–spike potentiation. (A) Representative recordings of population spike before and after application of 100 lM furosemide. Furosemide significantly increased the amplitude of PS. (B) The EPSP–spike (E–S) curve obtained from the representative recordings before and after the application of furosemide. Furosemide caused a leftward shift of E–S curve. (C) The summarized values of E50 in the E–S curve before and after the application of furosemide are shown in the histograms (n = 5). *P < 0.05.

involvement of KCC2 in weakening GABAergic inhibition and controlling circuit output during LTP. Weakening of GABAergic inhibition plays a critical role in regulating neuronal circuit plasticity and output [29]. Previously, much attention focused on the alteration of the amount of presynaptic transmitter release, the number and/or properties of synaptic receptor [30,31]. Interestingly, recent investigations evidenced that changes of postsynaptic Cl transporters play an important role in modifying GABAergic synaptic strength [11,18]. These results suggest that KCC2 might also be involved in the regulation of circuit flexibility. Activity-dependent downregulation of KCC2 was regarded as a general mechanism underlying GABAergic disinhibition in many diseases, such as ischemia, pain and epilepsy [14,19,20]. Although the change and contribution

of KCC2 is still unclear in LTP expression, increasing information has pointed to the potential effects of KCC2 in synaptic plasticity. Rivera et al. [17] evidenced that KCC2 shows an activity-dependent downregulation in hippocampus by high extracellular concentration of K+ ([K+]o) or low extracellular concentration of Mg2+ treatment. It’s well known that elevation of [K+]o is commonly applied to facilitate LTP [32]. Interestingly, high [K+]o was also regarded as a blocker of KCC2 and agitator of circuit activity [33,34]. These results suggest that inhibition of KCC2 may contribute to synaptic plasticity at least in hippocampus. In the present study, we found that KCC2 level was specifically decreased after LTP induction, while the amount of KCC2 was not altered in the process of LTD expression. Furthermore, a recent paper has well evidenced that function of KCC2 shows a stimulation

1214

W. Wang et al. / Biochemical and Biophysical Research Communications 343 (2006) 1209–1215

pattern-dependent property [18]. Consequently, these results suggest that a specific activity is necessary in regulating KCC2 function and level. On the other hand, activation of NMDAR during TBS is considered to be a critical mechanism for the formation of LTP [35]. In the present study, blockade of NMDAR not only prevented LTP induction also blocked the downregulation of KCC2. The data further indicated that KCC2 downregulation is a general process after LTP induction. Meanwhile, NMDAR-mediated Ca2+ influx may regulate the transcription and cause a reduction of KCC2 mRNA and/or protein which has been observed in neuronal injury or neurotoxic conditions [12,13,17]. Furthermore, the function of KCC2 can be downregulated by protein kinase C (PKC)-dependent phosphorylation [18]. Although, we evidenced that NMDAR-dependent downregulation of KCC2 was specifically accompanied with TBS-induced LTP formation, it would be very interesting to know whether other mechanisms other than the reduced expression of KCC2 also play a role for KCC2 regulation. Development-dependent upregulation of KCC2 is mainly distributed in membrane of mature neuron [11,18,20]. A recent paper evidenced that KCC2 primarily expressed in soma and dendrites [36]. Interestingly, GABAergic synapses also mainly located at perisomatic region of neuron [37]. The distribution pattern of KCC2 and GABAergic synapses provides the structure basis of controlling neuronal output. Specially, excitation and signal propagation initiated by axo-axonic cells is supported by the absence of the KCC2 in the axon [36]. Indeed, our unpublished data showed that inhibition of KCC2 significantly enhanced circuit activity in hippocampal slices (data not shown). Longterm potentiation accompanied EPSP–spike potentiation represent increased output of neuronal circuit [38]. In the present experiments, our data indicate that LTP-accompanied downregulation of KCC2 may be a potential mechanism underlying E–S potentiation. Additionally, using the KCC2 inhibitor furosemide, we further showed that the inhibition of KCC2 directly caused E–S potentiation, a process that is subsequent to the LTP formation. Downregulation of KCC2 results in a reduction of GABAergic inhibition through the elevation of the intracellular Cl level in postsynaptic neurons [14]. Moreover, previous studies have indicated that calcineurin-mediated downregulation of GABAergic inhibition through dephosphorylation of postsynaptic GABAAR contributes to the E–S potentiation in LTP [4]. Thus, our findings add a new dimension to understanding the complex modulation of hippocampal synaptic plasticity. We conclude that downregulation of KCC2 is specifically accompanied with TBS-induced LTP formation, which depends on activation of NMDAR. Inhibition of KCC2 directly increases circuit output. These results suggest that downregulation of KCC2 may contribute to the E–S potentiation following LTP via the reduction of the GABAergic inhibition. Further work is necessary to examine the

change of Cl in the process of LTP expression and its potential contribution to neuronal excitability. References [1] G. Buzsaki, Theta oscillations in the hippocampus, Neuron 33 (2002) 325–340. [2] R.C. Malenka, M.F. Bear, LTP and LTD: an embarrassment of riches, Neuron 44 (2004) 5–21. [3] D.J. Linden, The return of the spike: postsynaptic action potentials and the induction of LTP and LTD, Neuron 22 (1999) 661–666. [4] Y.M. Lu, I.M. Mansuy, E.R. Kandel, J. Roder, Calcineurinmediated LTD of GABAergic inhibition underlies the increased excitability of CA1 neurons associated with LTP, Neuron 26 (2000) 197–205. [5] H. Wigstrom, B. Gustafsson, Facilitated induction of hippocampal long-lasting potentiation during blockade of inhibition, Nature 301 (1983) 603–604. [6] C.A. Chapman, Y. Perez, J.C. Lacaille, Effects of GABA(A) inhibition on the expression of long-term potentiation in CA1 pyramidal cells are dependent on tetanization parameters, Hippocampus 8 (1998) 289–298. [7] R.A. Tomasulo, W.B. Levy, O. Steward, LTP-associated EPSP/spike dissociation in the dentate gyrus: GABAergic and non-GABAergic components, Brain Res. 561 (1991) 27–34. [8] V. Chevaleyre, P.E. Castillo, Heterosynaptic LTD of hippocampal GABAergic synapses: a novel role of endocannabinoids in regulating excitability, Neuron 38 (2003) 461–472. [9] W.E. Ghijsen, E. Besselsen, V. Geukers, W. Kamphuis, F.H. Lopes da Silva, Enhancement of endogenous release of glutamate and gamma-aminobutyric acid from hippocampus CA1 slices after in vivo long-term potentiation, J. Neurochem. 59 (1992) 482–486. [10] Y. Ben-Ari, R. Cossart, Kainate, a double agent that generates seizures: two decades of progress, Trends Neurosci. 23 (2000) 580–587. [11] C. Rivera, J. Voipio, J.A. Payne, E. Ruusuvuori, H. Lahtinen, K. Lamsa, U. Pirvola, M. Saarma, K. Kaila, The K+/Cl co-transporter KCC2 renders GABA hyperpolarizing during neuronal maturation, Nature 397 (1999) 251–255. [12] J. Nabekura, T. Ueno, A. Okabe, A. Furuta, T. Iwaki, C. ShimizuOkabe, A. Fukuda, N. Akaike, Reduction of KCC2 expression and GABAA receptor-mediated excitation after in vivo axonal injury, J. Neurosci. 22 (2002) 4412–4417. [13] H. Toyoda, K. Ohno, J. Yamada, M. Ikeda, A. Okabe, K. Sato, K. Hashimoto, A. Fukuda, Induction of NMDA and GABAA receptormediated Ca2+ oscillations with KCC2 mRNA downregulation in injured facial motoneurons, J. Neurophysiol. 89 (2003) 1353–1362. [14] J.A. Coull, D. Boudreau, K. Bachand, S.A. Prescott, F. Nault, A. Sik, P. De Koninck, Y. De Koninck, Trans-synaptic shift in anion gradient in spinal lamina I neurons as a mechanism of neuropathic pain, Nature 424 (2003) 938–942. [15] F. Galeffi, R. Sah, B.B. Pond, A. George, R.D. Schwartz-Bloom, Changes in intracellular chloride after oxygen-glucose deprivation of the adult hippocampal slice: effect of diazepam, J. Neurosci. 24 (2004) 4478–4488. [16] X. Jin, J.R. Huguenard, D.A. Prince, Impaired Cl extrusion in layer V pyramidal neurons of chronically injured epileptogenic neocortex, J. Neurophysiol. 93 (2005) 2117–2126. [17] C. Rivera, J. Voipio, J. Thomas-Crusells, H. Li, Z. Emri, S. Sipila, J.A. Payne, L. Minichiello, M. Saarma, K. Kaila, Mechanism of activity-dependent downregulation of the neuron-specific K–Cl cotransporter KCC2, J. Neurosci. 24 (2004) 4683–4691. [18] H. Fiumelli, L. Cancedda, M.M. Poo, Modulation of GABAergic transmission by activity via postsynaptic Ca2+-dependent regulation of KCC2 function, Neuron 48 (2005) 773–786. [19] N.S. Woo, J. Lu, R. England, R. McClellan, S. Dufour, D.B. Mount, A.Y. Deutch, D.M. Lovinger, E. Delpire, Hyperexcitability and

W. Wang et al. / Biochemical and Biophysical Research Communications 343 (2006) 1209–1215

[20]

[21]

[22]

[23]

[24]

[25]

[26] [27]

[28]

[29]

epilepsy associated with disruption of the mouse neuronal-specific K– Cl cotransporter gene, Hippocampus 12 (2002) 258–268. J.A. Payne, C. Rivera, J. Voipio, K. Kaila, Cation-chloride cotransporters in neuronal communication, development and trauma, Trends Neurosci. 26 (2003) 199–206. J. Gao, B. Duan, D.G. Wang, X.H. Deng, G.Y. Zhang, L. Xu, T.L. Xu, Coupling between NMDA receptor and acid-sensing ion channel contributes to ischemic neuronal death, Neuron 48 (2005) 635–646. U. Staubli, J. Scafidi, D. Chun, GABAB receptor antagonism: facilitatory effects on memory parallel those on LTP induced by TBS but not HFS, J. Neurosci. 19 (1999) 4609–4615. G. Capocchi, M. Zampolini, J. Larson, Theta burst stimulation is optimal for induction of LTP at both apical and basal dendritic synapses on hippocampal CA1 neurons, Brain Res. 591 (1992) 332– 336. J.A. Kauer, Blockade of hippocampal long-term potentiation by sustained tetanic stimulation near the recording site, J. Neurophysiol. 81 (1999) 940–944. J. Larson, D. Wong, G. Lynch, Patterned stimulation at the theta frequency is optimal for the induction of hippocampal long-term potentiation, Brain Res. 368 (1986) 347–350. U. Staubli, G. Lynch, Stable hippocampal long-term potentiation elicited by ‘theta’ pattern stimulation, Brain Res. 435 (1987) 227–234. A. Volianskis, M.S. Jensen, Transient and sustained types of longterm potentiation in the CA1 area of the rat hippocampus, J. Physiol. 550 (2003) 459–492. J.A. Payne, Functional characterization of the neuronal-specific K–Cl cotransporter: implications for [K+]o regulation, Am. J. Physiol. 273 (1997) C1516–C1525. P. Somogyi, G. Tamas, R. Lujan, E.H. Buhl, Salient features of synaptic organisation in the cerebral cortex, Brain Res. Brain Res. Rev. 26 (1998) 113–135.

1215

[30] J.M. Fritschy, C. Schweizer, I. Brunig, B. Luscher, Pre- and postsynaptic mechanisms regulating the clustering of type A gammaaminobutyric acid receptors (GABAA receptors), Biochem. Soc. Trans 31 (2003) 889–892. [31] D.M. Kullmann, A. Ruiz, D.M. Rusakov, R. Scott, A. Semyanov, M.C. Walker, Presynaptic, extrasynaptic and axonal GABAA receptors in the CNS: where and why? Prog. Biophys. Mol. Biol. 87 (2005) 33–46. [32] M.W. Fleck, A.M. Palmer, G. Barrionuevo, Potassium-induced longterm potentiation in rat hippocampal slices, Brain Res. 580 (1992) 100–105. [33] M.A. Woodin, K. Ganguly, M.M. Poo, Coincident pre- and postsynaptic activity modifies GABAergic synapses by postsynaptic changes in Cl transporter activity, Neuron 39 (2003) 807–820. [34] A. Semyanov, O. Godukhin, Epileptiform activity and EPSP–spike potentiation induced in rat hippocampal CA1 slices by repeated highK(+): involvement of ionotropic glutamate receptors and Ca(2+)/ calmodulin-dependent protein kinase II, Neuropharmacology 40 (2001) 203–211. [35] M.A. Lynch, Long-term potentiation and memory, Physiol. Rev. 84 (2004) 87–136. [36] J. Szabadics, C. Varga, G. Molnar, S. Olah, P. Barzo, G. Tamas, Excitatory effect of GABAergic axo-axonic cells in cortical microcircuits, Science 311 (2006) 233–235. [37] B. Chattopadhyaya, G. Di Cristo, H. Higashiyama, G.W. Knott, S.J. Kuhlman, E. Welker, Z.J. Huang, Experience and activity-dependent maturation of perisomatic GABAergic innervation in primary visual cortex during a postnatal critical period, J. Neurosci. 24 (2004) 9598– 9611. [38] G. Daoudal, Y. Hanada, D. Debanne, Bidirectional plasticity of excitatory postsynaptic potential (EPSP)–spike coupling in CA1 hippocampal pyramidal neurons, Proc. Natl. Acad. Sci. USA 99 (2002) 14512–14517.