Doi:10.1016/j.neulet.2004.02.040

Post-synaptic N-methyl-D-aspartate signalling in hippocampal neurons ofrat: spillover increases the impact of each spike in a short burst dischargeq Sergei E. Grebenyuk, Natasha A. Lozovaya, Timur S. Tsintsadze, Oleg A. Krishtal* Bogomoletz Institute of Physiology, 4, Bogomoletz St., 01024 Kiev, Ukraine High-frequency burst discharges in hippocampus typically consist of less than ten spikes fired at frequencies too high to be followed by a post-synaptic neuron. How significant are these numbers for synaptic signalling? We have measured the N-methyl-D-aspartate (NMDA)component of the excitatory post-synaptic current (EPSCNMDA) in hippocampal CA1 neurons of rat after burst discharge of variable duration.
The synaptic facilitation is accompanied by a slow-down of the EPSCNMDA which develops on a spike-to-spike basis. Consequently thecharge transferred by the after-burst EPSCNMDA is increased with each spike. The phenomenon is most probably due to the spillover-mediated recruitment of extrasynaptic NMDA receptors. In terms of post-synaptic signalling it dramatically increases the impact of eachspike in a short burst discharge.
q 2004 Elsevier Ireland Ltd. All rights reserved.
Keywords: N-Methyl-D-aspartate receptors; Hippocampus; CA3/CA1 synapses; Burst discharge; Extrasynaptic receptors; Spillover Natural spiking patterns in the brain are composed of this question, we analyzed a-amino-3-hydroxy-5 methyl- relatively short periods of high-frequency activity. From isoxazole-4-proprionic acid (AMPA) and an N-methyl-D- hippocampus through cortex, high-frequency burst dis- aspartate (NMDA) component of the excitatory post- charges are believed to be associated with information synaptic currents (EPSCs) in CA1 pyramidal neurons processing and memory consolidation. Specifically in elicited by short trains of high-frequency stimulation of hippocampus, they have two to nine action potentials fired at frequencies up to 200 Hz . How significant is every This study was carried out on 21-day-old Wistar rats spike within this small number? This question seems to be (WAG/GSto, Moscow, Russia). After rapid decapitation, rat especially important since at high frequencies of firing the brains were immediately transferred to a Petri dish with a post-synaptic neuron cannot follow pre-synaptic burst chilled (4 8C) solution of the following composition: 120 discharge with an equivalent number of spikes. To address mM NaCl, 5 mM KCl, 26 mM NaHCO3, 2 mM MgCl2 and20 mM glucose. The solution was constantly bubbled with a q Laudation in honour of Professor Zimmermann Great thanks to 95% O2/5% CO2 gas mixture to maintain pH 7.4. During the Professor Zimmermann from the senior author, Oleg A. Krishtal: I cannot pre-incubation and experiments, the slices (300 – 400 mM find a better word than ‘visionary’ to describe the Editorial activities of thick) were kept fully submerged in the extracellular Professor Zimmermann. At least on two occasions his openness to non- solution: 135 mM NaCl, 5 mM KCl, 26 mM NaHCO3, 1.5 paradigmal findings was crucial to the first publication on the ‘proton receptor and the ATP receptor in sensory neurons’ O.A. Krishtal, V.I.
Pidoplichko, A “receptor” for protons in small neurons of trigeminal bubbled with 95% O2/5% CO2) at 30 –31 8C. Picrotoxin ganglia: possible role in nociception. Neurosci. Lett., 24 (1981) 243 – 246 (25 – 50 mM) was added to the extracellular solution during O.A. Krishtal, S.M. Marchenko, V.I. Pidoplichko, Receptor for ATP in the the experiments to suppress the inhibitory activity of membrane of mammalian sensory neurones. Neurosci. Lett., 35 (1983) 41 – 45. Both papers dealt with unusual findings and this was a penetratingintuition of the Editor to say ‘Yes’. Subsequent history demonstrated that he Standard whole-cell patch clamp techniques were used to was right: the first paper resulted in the cloning of ASICs and in establishing record EPSCs from CA1 pyramidal neurons in situ in their role as nociceptors, while the second one contributed to the role of the response to stimulation of the Schaffer collateral/commis- P2X receptor family in sensory function.
sural pathway. To prevent the spread of electrical activity * Corresponding author. Tel.: þ 38-044-2532466; fax: þ 38-044- from area CA3, mini-slices were prepared by making a cut E-mail address: krishtal@serv.biph.kiev.ua (O.A. Krishtal).
orthogonal to the stratum pyramidale and extending to the 0304-3940/03/$ - see front matter q 2004 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.neulet.2004.02.040 S.E. Grebenyuk et al. / Neuroscience Letters 361 (2004) 60–63 mossy fibres layer. The intracellular solution for patchpipettes contained 100 mM CsF, 40 mM NaH2PO4, 10 mMHEPES-CsOH, and 10 mM Tris – Cl (pH 7.2). N-(2,6-Dimethyl-phenylcarbamoylmethyl)-triethylammonium bro-mide (QX-314) (2 – 3 mM) was routinely added to theintracellular solution to block voltage-gated sodium con-ductance. Patch pipettes were pulled from soft borosilicateglass on a two-stage horizontal puller. When fire-polishedand filled with the intracellular solution, they had aresistance of 2 – 3 MV. To visualize cell bodies of CA1pyramidal neurons, the stratum oriens and alveus wereremoved by a saline jet from a micropipette. Currents weredigitally sampled at 400 ms intervals by a 12-digit ADCboard, filtered at 3 kHz, and stored on a hard disk for furtheranalysis. Access resistance was monitored throughout theexperiments and ranged typically from 6 to 9 MV. When theaccess resistance changed by more than 25% during theexperiment, the data were discarded. To stimulate a Schaffercollateral/commissural pathway, a bipolar Ni/Cr electrodewas positioned on the surface of the slice. Current pulses(10 – 100 mA) of 0.1 – 1 ms duration were delivered throughthe isolated stimulator HG 203 (Hi-Med, London, UK) at0.066 – 0.2 Hz.
Sodium bicarbonate and CsF were obtained from Merck Fig. 1. Facilitation of EPSC by short trains of stimulation is accompanied (Darmstadt, Germany); 4-AP, lidocaine and picrotoxin were by a selective slow-down of the NMDA component. (A) (Left) purchased from RBI (Natick, MA); 6-nitro-7-sulphamoyl- Pharmacologically isolated EPSCAMPA evoked by a single stimulus applied benzo[ f ]quinoxalin-2,3-dione (NBQX) was obtained from to Schaffer collaterals is superimposed with the EPSCAMPA evoked by a ten Tocris Cookson (Bristol, UK). All other chemicals were pulses long (200 Hz) train (n ¼ 7). Here and below: stimulation protocols are schematically represented over the traces; the insets demonstratenormalized time courses of the currents. (Right) The same for EPSC When measured after a short train of stimuli delivered at (n ¼ 30). (B) The kinetics of EPSCNMDA evoked by a single stimulus does high frequency, kinetics of the AMPA component of EPSC not depend on the strength of a single stimulus (left), but becomes stimulus- are the same as after a single stimulus, while the NMDA- dependent after seven pulses at 200 Hz (right) (n ¼ 10). The stimulus intensities varied by a factor of six. (C) (Left) The slowdown of EPSC goes a dramatic slow-down The changes in the is not determined solely by the current amplitudes: the traces from (B)elicited by the largest single stimulus (*) and by the smallest stimulus in the NMDA kinetics were quantified by normalizing the train (**) have equal amplitudes (superimposed, not normalized). The charge transfer with the peak amplitude of the EPSCNMDA kinetics of the post-train EPSCNMDA is evidently slower. (Right) Slow- (measured as the mean over a 10 ms window around the down of EPSCNMDA decay under high-frequency stimulation is clearly seen peak). A larger charge transfer corresponds to slower decay at positive voltage þ 20 mV (n ¼ 4), when the voltage-dependent kinetics and vice versa. The normalized charge transfer of conductance is inactivated and voltage-dependent Mg2þ block of NMDAchannels is removed. The small current elicited by a single pulse is NMDA was 355 ^ 32% of the same parameter for normalized (*) to the peak value of current elicited by seven pulses at 200 NMDA induced by a single stimulus (EPSCNMDA) Hz. (D) Theoretical changes in the EPSCNMDA kinetics due to the change in the current amplitude. The plot shows the dependence of charge transferred on the EPSCNMDA amplitude measured in the soma attributed to inadequate voltage clamp. A biophysical model of CA1 pyramidal cells was adapted to ourexperimental conditions in order to estimate the extent to strates the responses of the same amplitude obtained after a which the quality of the voltage clamp affects the EPSC single stimulus and a burst stimulation. Their kinetics are decay. The density of synapses in the model of CA1 neuron obviously different. Therefore, at least qualitatively, we can was varied and the EPSCNMDA amplitude, as well as the rely on the experimental data: it is not the increase in the changes in EPSCNMDA kinetics, were analyzed. amplitude of the current that solely determines the slow- demonstrates the theoretical relationship between the down of EPSCNMDA, though, for a given number of stimuli, amplitude and kinetics of EPSCNMDA. The experimentally the larger the stimulus applied, the greater this effect obtained changes in EPSCNMDA amplitude (, 1.5-fold increase) can be associated with , 5% changes in the The behaviour of EPSC components after the train ( calculated charge transfer vs. 355% changes in the remarkably resembles the findings previously made in experimentally detected charge transfer. demon- conditions of enhanced transmitter release: the AMPA S.E. Grebenyuk et al. / Neuroscience Letters 361 (2004) 60–63 component has unaltered kinetics, while the NMDA NMDA receptor saturation by single quanta and component slows down and acquires stimulus-dependent provide a basis for synaptic integration.
kinetics This phenomenology is consistent with the Thus, the synaptic facilitation by a short train of spillout of Glu from the synaptic cleft resulting in a cross- stimulation imitating the natural pattern of burst dis- talk between neighbouring synapses and/or charge leads to the increased impact of spillover- activation of extrasynaptic NMDA receptors If this activated NMDA receptors. For a given number of picture is correct, the enhanced release of Glu should have stimuli in the train, the higher the frequency used ( the following consequences: (i) the contribution of extra- synaptic receptors should increase; (ii) these receptors should ‘see’ considerably lower concentrations of Glu as demonstrates the spike-to-spike changes in the compared to the case of the receptors in the post-synaptic EPSCNMDA. The kinetics becomes notably altered starting already from the second stimulus. The charge transferred by The efficacy of low-affinity competitive NMDA receptor EPSCNMDA allows the estimation of the dependence of antagonist D-amino adipate (D-AA) is known to depend on Ca2þ entering the cell on the number of spikes in the pre- the synaptic concentration of glutamate It has been used synaptic burst discharge When compared with a to distinguish receptors activated within active synapses hypothetical dependence for the case of kinetics which from those activated by spillover demonstrates would remain unaltered by the burst discharge (only the the experiments in which the prolonged after-train EPSCNMDA was elicited in the presence of D-AA. The steeper. Therefore, the information on the pre-synaptic blocker leaves practically unaltered the kinetics of activity (the number of spikes and of activated synapses) is EPSCNMDA in the control, but markedly speeds up thedecay of EPSCNMDA induced by the train. Taken as ameasure of the change in the kinetics, normalized chargetransfer produced by the EPSCtrain D-AA was 65 ^ 2% of control, while the correspondingvalue for the EPSCsingle NMDA was 88 ^ 8% (P , 0:005, n ¼ 4).
Preferential inhibition of a slower component of theEPSCNMDA suggests that the NMDA receptors responsiblefor this component are activated by lower [Glu] than thosecontributing to the peak of the current.
We have found an indication of a dramatic increase in the transmitter concentration acting at the peak of after-train EPSCNMDA. Inhibition of current by 50 mM D-AAin (80 ^ 18%, n ¼ 4) at the peak of EPSCNMDA recordedafter a train of seven stimuli (The 1.25-folddecrease in EPSCNMDA amplitude observed in the D-AAexperiments should result in a 4% decrease in thetheoretical normalized charge transfer, whereas the actualchanges comprised 23%. The most straightforwardexplanation of this phenomenon is the induction of multivesicular transmitter release in the course of high- NMDA depends on the frequency of stimulation and on the number of spikes in a train. (A) The slow-down of EPSCNMDA frequency firing . Such a possibility can serve as decay depends on the frequency of stimulation within the train. The a functional ‘justification’ for the lack of synaptic responses to seven stimuli at 5, 10, 20, 50 and 100 Hz are superimposed onthe left graph. Their normalized decays are on the right graph. (B) Theslow-down of EPSCNMDA on a spike-to-spike basis. (Left) Family ofEPSCNMDA evoked by the progressively increased number of pulses (one tonine) in the train (200 Hz). Holding voltage 2 45 mV. (Right)Corresponding traces are normalized and superimposed. (C) (Left) Thecharge transferred by after-burst EPSCNMDA depends on the number ofspikes (one to seven) in the train. The data obtained for five neuronsstimulated at 200 Hz (squares) are compared with the values obtained for Fig. 2. D-AA accelerates after-burst EPSCNMDA. Block of EPSCNMDA by the hypothetical case of unaltered kinetics (only amplitude facilitation is D-AA induced by a single pulse (left) and by seven pulses at 200 Hz (right).
accounted for; triangles). (Right) The charge transferred by EPSCNMDA, Holding voltage 2 45 mV. The insets demonstrate that D-AA does not alter normalized to the peak amplitude. The dependence saturates after the 5th to the kinetics of the current elicited by a single pulse, but speeds up the after- 6th stimulus. In the case of unaltered kinetics this dependence would S.E. Grebenyuk et al. / Neuroscience Letters 361 (2004) 60–63 synaptic transmission, in: C. Koch, I. Segev (Eds.), Methods in (right) demonstrates that the gradual change in the Neuronal Modeling, MIT Press, Cambridge, MA, 1998, pp. 1 – 25.
[6] J.S. Diamond, Neuronal glutamate transporters limit activation of EPSCNMDA kinetics is saturated by the 5th to 6th stimulus.
NMDA receptors by neurotransmitter spillover on CA1 pyramidal The experiments with Ca2þ imaging reveal that NMDA cells, J. Neurosci. 21 (2001) 8328 – 8338.
receptors play a leading role in creating the post-synaptic [7] E.R. Kandel, W.A. Spencer, Electrophysiology of hippocampal Ca2þ signal and in controlling plasticity Spillover neurons. II. Afterpotentials and repetitive firing, J. Neurophysiol. 24 seems to play an important role in shaping this signal by [8] Y. Kovalchuk, J. Eilers, J. Lisman, A. Konnerth, NMDA receptor- enhancing its dependence on the number of spikes in the mediated subthreshold Ca(2 þ ) signals in spines of hippocampal neurons, J. Neurosci. 20 (2000) 1791 – 1799.
[9] J.E. Lisman, Bursts as a unit of neural information: making unreliable synapses reliable, Trends Neurosci. 20 (1997) 38 – 43.
[10] N.A. Lozovaya, M.V. Kopanitsa, Y.A. Boychuk, O.A. Krishtal, Enhancement of glutamate release uncovers spillover-mediatedtransmission by N-methyl-D-aspartate receptors in the rat hippo- This work was supported by the Wellcome Trust and campus, Neuroscience 91 (1999) 1321 – 1330.
[11] Z.F. Mainen, R. Malinow, K. Svoboda, Synaptic calcium transients in single spines indicate that NMDA receptors are not saturated, Nature399 (1999) 151 – 155.
[12] A.K. McAllister, C.F. Stevens, Nonsaturation of AMPA and NMDA receptors at hippocampal synapses, Proc. Natl. Acad. Sci. USA 97(2000) 6173 – 6178.
[1] N. Arnth-Jensen, D. Jabaudon, M. Scanziani, Cooperation between [13] M. Migliore, D.A. Hoffman, J.C. Magee, D. Johnston, Role of an A- independent hippocampal synapses is controlled by glutamate uptake, type K þ conductance in the back-propagation of action potentials in Nat. Neurosci. 5 (2002) 325 – 331.
the dendrites of hippocampal pyramidal neurons, J. Comput.
[2] F. Asztely, G. Erdemli, D.M. Kullmann, Extrasynaptic glutamate spillover in the hippocampus: dependence on temperature and the role [14] T.G. Oertner, B.L. Sabatini, E.A. Nimchinsky, K. Svoboda, of active glutamate uptake, Neuron 18 (1997) 281 – 293.
Facilitation at single synapses probed with optical quantal analysis, [3] J.D. Clements, R.A. Lester, G. Tong, C.E. Jahr, G.L. Westbrook, The Nat. Neurosci. 5 (2002) 657 – 664.
time course of glutamate in the synaptic cleft, Science 258 (1992) [15] D.A. Rusakov, D.M. Kullmann, Extrasynaptic glutamate diffusion in the hippocampus: ultrastructural constraints, uptake, and receptor [4] R. Conti, J. Lisman, The high variance of AMPA receptor- and activation, J. Neurosci. 18 (1998) 3158 – 3170.
NMDA receptor-mediated responses at single hippocampal synapses: [16] S.S. Suzuki, G.K. Smith, Burst characteristics of hippocampal evidence for multiquantal release, Proc. Natl. Acad. Sci. USA 100 complex spike cells in the awake rat, Exp. Neurol. 89 (1985) 90 – 95.
[17] G. Tong, C.E. Jahr, Multivesicular release from excitatory synapses of [5] A. Destexhe, Z.F. Mainen, T.J. Sejnowski, Biophysical models of cultured hippocampal neurons, Neuron 12 (1994) 51 – 59.

Source: http://www.deptcm.kiev.ua/tt/2.pdf