- Email: [email protected]
Electrophysiological properties of rat subicular neurons in vitro
ELSEVIER
Neuroscience Letters 220 (1996) 41-44
NHHSCIgiC[ LETTtRS
Electrophysiological properties of rat subicular neurons in vitro J. Behr*, R.M. Empson, D. Schmitz, T. Gloveli, U. Heinemann Institute of Physiology at the Charitd, Department of Neurophysiology, Humboldt University Berlin, Tucholskystrasse 2, D-lOll 7 Berlin, Germany
Received 16 October 1996; revised version received 5 November1996; accepted 6 November 1996
Abstract
The electrophysiological properties of 46 bursting cells and 39 regular firing cells were studied in the subiculum of rat combined hippocampal-entorhinal coilex slices. In bursting cells we found a significantly higher resting membrane potential than in regular firing cells. Upon hyperpolarization both cell types expressed a delayed inward rectification with a subsequent afterdepolarization. While in regular firing cells longer lasting depolarizing current injection caused a train of action potentials with a rather marked decline of discharge frequency, bursting cells displayed only little frequency accommodation. Regular firing cells usually displayed a fast and a slow afterhyperpolarization following a train of action potentials, while bursting neurons present only a slow afterhyperpolarization. Copyright © 1996 Elsevier Science Ireland Ltd. Keywords: Subiculum; Intracellular recording; Bursting cells; Regular firing cells; Accommodation; Afterhyperpolarization
The subiculum receives strong input from area CA1 [1,4,14] and the entorhinal cortex (EC) [12,16]. It sends efferents to other region:~ in the subicular complex, to the deep and superficial layers of the EC and a variety of distant cortical and subcoritcal structures [16-18] and is therefore the main output structure of the hippocampus (HC). Previous studies investigated the electrophysiological properties of subicular neurons and divided them into two cell types based on their response to depolarizing current injections; bursting and regular firing cells [6,8,13,15]. We were interested, whether features such as spike frequency accommodation and afterhyperpolarization were similar in both cell classes. The experiments were performed on 62 horizontal slices containing the entorhinal cortex, the subiculum and the hippocampal formation obtained from 180-230 g Wistar rats. The rats were decapitated under deep ether anesthesia, the brains quickly removed and 400/xm thick slices were prepared with a Cambden vibroslicer (Loughborough, UK). The slices were transferred into an interface chamber continuously perfused with an aerated (95% 02, 5% CO2), prewarmed (34°C) artificial cerebrospinal fluid (ACSF) containing in raM: NaC1 124, NaH2PO4 1.25, * Corresponding author. Tel.: +49 30 28026640; fax: +49 30 28026669; e-mail: [email protected]
NaHCO3 26, KC1 3, CaC12 1.6, MgSO4 1.8, glucose 10, pH 7.4. Membrane potential (MP) recordings from ventral subicular neurons were made exclusively in the cell band in extension of the pyramidal cell layer in area CA1 within close proximity to the perforating fibres of the perforant path using sharp microelectrodes (resistance 50-80 Mfi) filled with 2.5 M K-acetate and a Neurodata IR 183 amplifier (Neurodata Instruments Corp., NY, USA) or a SEC10L amplifier (npi Instruments, Tamm, Germany) in conventional bridge mode. Cells with membrane potentials more negative than - 5 0 mV and overshooting action potentials were accepted. Signals were filtered at 3 kHz, sampled and collected using a TIDA interface card (Batelle, Frankfurt, Germany). All data were analyzed off line using TIDA software (HEKA, Lambrecht/Pfalz, Germany). Statistical evaluation was performed by calculating means + SD of the mean (AXUM, Axum, Seattle, USA) and applying a student t-test (Sigmaplot, Jandel, Corte Madera, USA). In this study stable intracellular recordings were obtained from 85 subicular neurons. Nearly half of the cells (n = 39) corresponded to regular firing neurons as described in neocortex [3] and in area CA1 [10, 11]. These cells had an average resting membrane potential of -63.7 + 5.8 mV and an input resistance of
0304-3940/96/$12.00 Copy~ght© 1996 Elsevier Science Ireland Ltd. All rights reserved PII S0304-3940(96) 13242-0
42
J. Behr et al. / Neuroscience Letters 220 (1996) 41-44
44.5 + 14.5 Mfl determined from responses to negative current pulses (-0.01 nA, 100 ms duration) applied at resting membrane potential. The mean threshold for generation of action potentials measured with threshold depolarizing current injection was -53.3 + 8.6 mV. The action potential amplitude defined as the difference between resting membrane potential and peak potential was 77.6 + 10.1 mV. The duration of an action potential determined at half amplitude was 0.70 + 0.12 ms (n = 38). Application of short depolarizing current pulses (20-40 ms) from MP o f - 8 0 to - 7 0 mV caused above threshold a single action potential (Fig. 1A). Upon longer lasting depolarizing current injections (800 ms, 0.05-2.0 nA) these cells displayed trains of action potentials with a rather marked decline of discharge frequency over time (Fig. 1B). The first interval between action potentials had a mean duration of 64.5 + 37.5 ms and the last interval of 159.0 + 89.3 ms (8-10 action potentials/pulse). Following induction of 5 - 6 action potentials during an 100 ms depolarizing current injection all these cells displayed a fast afterhyperpolarization (AHP) that peaked 24.0 + 10.3 ms after the end of the current pulse. The amplitude of the AHP was -3.0 + 1.4 mV (n = 14). Thirteen of these ceils showed also a slow afterdepolarization o f - 1 . 5 + 1.3
mV that was measured at its peak or, if the afterdepolarization displayed a gradual decay, 300 ms after the end of the current injection (Fig. 1C). Upon injections of hyperpolarizing current (100 ms) applied at rest, all regular firing cells displayed an inward rectification ('sag') and a slow afterdepolarization at the end of the current pulses that increased with rising injection of hyperpolarizing current (Fig. 1E). In some neurones the afterdepolarization reached threshold and consequently elicited an action potential, but never a bursting discharge. Fig. 1D shows that both, the 'sag' potential and the slow afterdepolarization decreased, when the MP was brought to more negative levels (15-20 mV below rest). The second group of subicular neurons (n = 46) responded to depolarizing current pulses with a brief burst of 2 - 3 action potentials. These cells showed passive membrane characteristics similar to those of regular firing neurons. The cells of this group had a mean resting membrane potential o f - 6 7 . 0 + 6.8 mV significantly different from regular firing cells (P < 0.05). Input resistance of 41.3 + 15.9 Mfl, the average of the action potential amplitudes of 77.5 + 8.4 mV, the action potential duration of 0.73 + 0.12 ms and mean threshold o f - 5 5 . 2 + 8.0 mV (n = 37) were
C
A
25 mV
25 mV
100ms
500ms 0.3nA
D
E (nA)
AI
-
-77mV~ - 87mV~
FL
1.4nA
2O
- 67rnV
•
5 mV 500ms
1
1
~
0.2 0.4
-20 -10 --
25mY
-30
50 ms -40
- - I
1.0nA
Fig. 1. Electrophysiological properties of subicular regular firing neurons. (A) Response to intracellular current injection at resting membrane potential. Pulse duration, 40 ms; current amplitudes, 0.05-0.3 nA, 0.05 nA increments. (B) Firing pattern and accommodation property near threshold following intracellular current injection. Pulse duration, 800 ms; current amplitude, 0.3 nA. Note the marked decline of discharge frequency over time. (C) Slow and fast afterdepolarization of regular firing cell following induction of 5-6 action potentials during a 100 ms depolarizing current pulse. (D) Voltage responses to intracellularly injected current pulses, voltage dependency of inward rectification and anodal break response. Pulse duration, 100 ms; current amplitudes, -1.0 to -0.2 nA, 0.2 nA steps. Note, (1) Time-dependent inward rectification and depolarizing afterpotential which increase with higher hyperpolarizing current injections, (2) at a membrane potential 15-20 mV more negative than the resting membrane potential inward rectification and anodal break response disappeared. (E) Mean plot of the current-voltage relationship (I-V) measured at the peak (0) and the 'steady state' (&) of the hyperpolarizing and depolarizing voltage deflections and at the peak (B) of the depolarizing afterpotential following 100 ms current pulses.
J. Behr et al. / Neuroscience Letters 220 (1996) 41-44
B
A
C
25 mV
100ms
0.2hA
L- . . . . . 500ms
j
]
E
D -65 mV ~
-75mV .
- 82 mV ~_~
~
25 mV
5my
0.25nA
I 1.5 nA
2O
~
~
43
I {hA)
.
-
~
~
1
~
~
-100.2 0.4
25 mV -,3O -40
Fig. 2. Electrophysiological ploperties of subicular bursting neurons. (A) Response to intracellular current injection at resting membrane potential. Pulse duration, 40 ms; current amplitudes, 0.05-0.2 nA, 0.05 nA increments. Note, (1) Short depolarizing current pulse elicited a self-sustained burst discharge of two action potentials which continued after the end of the pulse, (2) burst is followed by a pronounced hyperpolarizing afterpotential. (B) Firing pattern and accommodation property near threshold following intracellular current injection. Pulse duration, 800 ms; current amplitude, 0.25 nA. Note, Bursting cell displayed little frequency accommodation over time. (C) Afterdepolarization of bursting cell following induction of a burst and 3-4 action potentials during a 100 ms depolarizing current pulse. Note that in contrast to regular firing cells this cell type does not present a slow afterdepolarization. (D) Voltage responses to intracell~alarly injected current pulses, voltage dependency of inward rectification and anodal break response. Pulse duration, 100 ms; current amplitudes, -1.0 to -0.2 nA, 0.2 nA steps. Note, (1) Time-dependent inward rectification and depolarizing afterpotential which increase with higher hyperpolarizing currenl injections, (2) at a membrane potential 15-20 mV more negative than the resting membrane potential inward rectification and anodal break response disappeared. (E) Mean plot of the current-voltage relationship (l-V) measured at the peak (@) and the 'steady state' (A) of the hyperpolarizing and depolarizing voltage deflections and at the peak (D) of the depolarizing afterpotential (anodal break response) following 100 ms current pulses.
not significantly different from regular firing cells (P >
0.05). Application of short depolarizing current pulses (20-40 ms) elicited in contrast to regular firing cells a self-sustained burst discharge of 2 - 3 action potentials which continued after the end of the pulse (Fig. 2A). The burst was followed by a pronounced negative afterpotential of --6.5 + 3.1 mV that peaked 29.3 + 7.2 ms after termination of the depolarizing cun'ent pulse (n = 7). Depolarization above threshold for 800 ; m s (0.05-2.0 nA) caused an initial burst of action potentials followed by an afterhyperpolarization and then by a train of single action potentials with a frequency depending on the magnitude of the depolarization (Fig. 2B). In contrast to regular firing cells bursting cells displayed little frequency accommodation. The interval between the first and second action potential following the burst was 83.6 + 30.6 ms and between the before last and the last action potential 87.7 + 33.6 ms (8-10 action potentials/pulse). Bursting cells displayed an afterhyperpolarization of -3.0 + 1.3 mV that peaked 22.5 + 5.0 ms after the end of a 100 ms depolarizing current pulse that induced a burst and 3 - 4 action potentials. The amplitude
of this fast afterhyperpolarization showed no difference to regular firing cells (P < 0.05). Twelve of 14 investigated bursting neurons did not present a slow afterdepolarization in contrast to regular firing cells (Fig. 2C) (P = 0.001). In response to hyperpolarizing current pulses bursting cells displayed also a 'sag' potential and a slow afterdepolarization that could be sufficient to elicit action potentials or burst discharges. Both could be blocked bY hyperpolarizing background currents that hyperpolarized the cell by 15-20 mV below resting potential (Fig. 2D). The dependence of the 'sag' potential and the slow afterdepolarization on membrane potential was similar to that of the regular firing neurons (Fig. 2E). The present study confirms a high incidence of bursting cells in the subiculum. We could show that beyond their firing pattern these cell type present distinct electrophysiological differences from regular firing cells. Like in previous studies the major division of subicular neurons was into burster and regular firing cells. Bursting could be induced either by depolarizing current pulses or by afterdepolarizations following the end of hyperpolarizing current injections. The burst generating mechanisms
44
J. Behr et al. / Neuroscience Letters 220 (1996) 41-44
underlying hyperpolarization induced bursting may be different from that following depolarization. Hyperpolarization induced bursting was previously shown to be sensitive to tetrodotoxin (TTX) [8] while depolarization induced bursting in the same type of neurons seemed to be insensitive to TTX [13]. The incidence of burster cells was in our study somewhat smaller than reported by Mattia et al. [8], Stewart and Wong [13], Mason [6] and Taube [15]. The incidence of burster cells in those studies was 60100%. As bursting cells can show single-spiking behavior when their membrane is depolarized, we always examined the firing patterns from a membrane potential of about -80 to -70 mV in order to avoid false classification. Differences in cell properties may also result from previous exposure to neuromodulators. Preliminary experiments have revealed that these neurons do not change their discharge behavior after previous exposure to 5-HT (Behr et al., submitted). Alternatively, spontaneous transmitter release may also influence discharge behavior. In our hands the discharge mode of a given neuron remained unaltered, when glutamate and GABA receptors were blocked. The difference in the incidence of burster cells may thus vary between subfields of the subiculum and along the dorso-ventral axis of the hippocampus. Previously reports have shown such variations in area CA1 [7]. Both cell types expressed upon hyperpolarization a delayed inward rectification with a subsequent afterdepolarization as has been previously reported [8,13,15]. The 'sag' potential was previously shown to be blocked by Cs ÷ [8,13] suggesting that it is dependent on an inward rectifying current consisting of a mixed Na÷/K ÷ conductance (IQ or IH) or an K ÷ inward rectifier. Indeed, activation of these currents by constant depolarization let the delayed inward rectification disappear. Besides the common intrinsic characteristics bursting and regular firing cells differ significantly with regard to accommodation properties and afterhyperpolarization. The lack of accommodation in bursting cells might be related to the absence of a slow AHP. The fast repolarization in bursting cells following an action potential facilitates the generation of subsequent action potentials whereas in regular firing cells Ca2÷-dependent slow AHPs that increase during a train of action potentials, might cause the marked accommodation. In contrast, bursting cells of the neocortex may present a slow AHP and during depolarizing current pulses their spike frequency does adapt [9]. Why does the subiculum contain a relatively extensive subpopulation of bursting cells with distinct intrinsic properties? An essential function of bursting cells is the amplification of neural signals [5]. As the subiculum is one of the important output stations for the hippocampal complex, its subpopulation of bursting neurons might serve as an amplifier facilitating the processing of information. This hypothesis may be confirmed by the fact that bursting cells do not display a marked accommodation. Above that
the hippocampal-entorhinal complex is known to be an epilepsy-prone area. It is known that the subiculum participates strongly in ictaform activity [2]. Therefore subicular bursting cells may support the generation and spread of convulsive activity. This work was supported by a grant from a DFG Innovationskolleg INK 21/Al-1 and the BMBF. We are indebted to H. Tetsch and A. Dtierkop for technical assistance in the experiments. [1] Amaral, D.G., Dolorfo, C. and Alvarez-Royo, P., Organization of CA1 projections to the subiculum: a PHA-L analysis in the rat, Hippocampus, 1 (1991)415-436. [2] Behr, J. and Heinemann, U., Low Mg 2÷ induced epilptiform activity in the subiculum before and after disconnection from rat hippocampal and entorhinal cortex slices, Neurosci. Lett., 205 (1996) 25-28. [3] Connors, B.W., Gutnick, M.J. and Prince, D.A., Electrophysiological properties of neocortical neurons in vitro, J. Neurophysiol., 48 (1982) 1302-1320. [4] Finch, D.M. and Babb, T.L., Demons~ation of caudally directed hippocampal efferents in the rat by intracellular injection of horseradish peroxidase, Brain Res., 214 (1981) 405-410. [5] Hablitz, J.J. and Johnston, D., Endogenous nature of spontaneous bursting in hippocampal pyramidal neurons, Cell. Mol. Neurobiol., 1 (1981) 325-333. [6] Mason, A., Electrophysiology and burst-firing of rat subicular pyramidal neurons in vitro: A comparison with area CA1, Brain Res., 600 (1993) 174-178. [7] Masukawa, L.M., Benardo, L.S. and Prince, D.A., Variations in electrophysiological properties of hippocampal neurons in different subfields, Brain Res., 242 (1982) 341-344. [8] Mattia, D., Hwa, G.G. and Avoli, M., Membrane properties of rat subicular neurons in vitro, J. Neurophysiol., 70 (1993) 1244-1248. [9] McCormick, D.A., Connors, B.W., Lighthall, J.W. and Prince, D.A., Comparative electrophysiology of pyramidal and sparsely spiny steUate neurons of the neocortex, J. Neurophysiol., 54 (1985) 782-806. [10] S c h w ~ o i n , P.A., Characteristics of CA1 neurons recorded intracellularly in the hippocampal 'in vitro' slice preparation, Brain Res., 85 (1975) 423-436. [11] Schwartzkroin, P.A., Further characteristics of hippocampal CA1 cells in vitro, Brain Res., 128 (1977) 53-68. [12] Steward, O. and Scoville, S.A., Cells of origin of entorhinal cortical afferents to the hippocampus and fascia dentata of the rat, J. Comp. Neurol., 169 (1976) 347-370. [13] Stewart, M. and Wong, R.K., Intrinsic properties and evoked responses of guinea pig subicular neurons in vitro, J. Neurophysiol., 70 (1993) 232-245. [14] Tamamaki, N. and Nojyo, Y., Disposition of the slab-like modules formed by axon branches originating from single CA1 pyramidal neurons in the rat hippocampus, J. Comp. Neurol., 291 (1990) 509-519. [15] Taube, J.S., Electrophysiological properties of neurons in the rat subiculum in vitro, Exp. Brain Res., 96 (1993) 304-318. [16] Witter, M.P., Organization of the entorhinal-hippocampal system: a review of current anatomical data, Hippocampus, 3 (1993) 33-44. [17] Witter, M.P. and Groenewegen, H.J., The subiculum: cytoarchitectonically a simple structure, but hodologically complex, Prog. Brain Res., 83 (1990) 47-58. [18] Witter, M.P., Ostendorf, R.H. and Groenewegen, H.J., Heterogeneity in the dorsal subiculum of the rat: distinct neuronal zones project to different cortical and subcortical targets, Eur. J. Neurosei., 2 (1990) 718-725.
Neuroscience Letters 220 (1996) 41-44
NHHSCIgiC[ LETTtRS
Electrophysiological properties of rat subicular neurons in vitro J. Behr*, R.M. Empson, D. Schmitz, T. Gloveli, U. Heinemann Institute of Physiology at the Charitd, Department of Neurophysiology, Humboldt University Berlin, Tucholskystrasse 2, D-lOll 7 Berlin, Germany
Received 16 October 1996; revised version received 5 November1996; accepted 6 November 1996
Abstract
The electrophysiological properties of 46 bursting cells and 39 regular firing cells were studied in the subiculum of rat combined hippocampal-entorhinal coilex slices. In bursting cells we found a significantly higher resting membrane potential than in regular firing cells. Upon hyperpolarization both cell types expressed a delayed inward rectification with a subsequent afterdepolarization. While in regular firing cells longer lasting depolarizing current injection caused a train of action potentials with a rather marked decline of discharge frequency, bursting cells displayed only little frequency accommodation. Regular firing cells usually displayed a fast and a slow afterhyperpolarization following a train of action potentials, while bursting neurons present only a slow afterhyperpolarization. Copyright © 1996 Elsevier Science Ireland Ltd. Keywords: Subiculum; Intracellular recording; Bursting cells; Regular firing cells; Accommodation; Afterhyperpolarization
The subiculum receives strong input from area CA1 [1,4,14] and the entorhinal cortex (EC) [12,16]. It sends efferents to other region:~ in the subicular complex, to the deep and superficial layers of the EC and a variety of distant cortical and subcoritcal structures [16-18] and is therefore the main output structure of the hippocampus (HC). Previous studies investigated the electrophysiological properties of subicular neurons and divided them into two cell types based on their response to depolarizing current injections; bursting and regular firing cells [6,8,13,15]. We were interested, whether features such as spike frequency accommodation and afterhyperpolarization were similar in both cell classes. The experiments were performed on 62 horizontal slices containing the entorhinal cortex, the subiculum and the hippocampal formation obtained from 180-230 g Wistar rats. The rats were decapitated under deep ether anesthesia, the brains quickly removed and 400/xm thick slices were prepared with a Cambden vibroslicer (Loughborough, UK). The slices were transferred into an interface chamber continuously perfused with an aerated (95% 02, 5% CO2), prewarmed (34°C) artificial cerebrospinal fluid (ACSF) containing in raM: NaC1 124, NaH2PO4 1.25, * Corresponding author. Tel.: +49 30 28026640; fax: +49 30 28026669; e-mail: [email protected]
NaHCO3 26, KC1 3, CaC12 1.6, MgSO4 1.8, glucose 10, pH 7.4. Membrane potential (MP) recordings from ventral subicular neurons were made exclusively in the cell band in extension of the pyramidal cell layer in area CA1 within close proximity to the perforating fibres of the perforant path using sharp microelectrodes (resistance 50-80 Mfi) filled with 2.5 M K-acetate and a Neurodata IR 183 amplifier (Neurodata Instruments Corp., NY, USA) or a SEC10L amplifier (npi Instruments, Tamm, Germany) in conventional bridge mode. Cells with membrane potentials more negative than - 5 0 mV and overshooting action potentials were accepted. Signals were filtered at 3 kHz, sampled and collected using a TIDA interface card (Batelle, Frankfurt, Germany). All data were analyzed off line using TIDA software (HEKA, Lambrecht/Pfalz, Germany). Statistical evaluation was performed by calculating means + SD of the mean (AXUM, Axum, Seattle, USA) and applying a student t-test (Sigmaplot, Jandel, Corte Madera, USA). In this study stable intracellular recordings were obtained from 85 subicular neurons. Nearly half of the cells (n = 39) corresponded to regular firing neurons as described in neocortex [3] and in area CA1 [10, 11]. These cells had an average resting membrane potential of -63.7 + 5.8 mV and an input resistance of
0304-3940/96/$12.00 Copy~ght© 1996 Elsevier Science Ireland Ltd. All rights reserved PII S0304-3940(96) 13242-0
42
J. Behr et al. / Neuroscience Letters 220 (1996) 41-44
44.5 + 14.5 Mfl determined from responses to negative current pulses (-0.01 nA, 100 ms duration) applied at resting membrane potential. The mean threshold for generation of action potentials measured with threshold depolarizing current injection was -53.3 + 8.6 mV. The action potential amplitude defined as the difference between resting membrane potential and peak potential was 77.6 + 10.1 mV. The duration of an action potential determined at half amplitude was 0.70 + 0.12 ms (n = 38). Application of short depolarizing current pulses (20-40 ms) from MP o f - 8 0 to - 7 0 mV caused above threshold a single action potential (Fig. 1A). Upon longer lasting depolarizing current injections (800 ms, 0.05-2.0 nA) these cells displayed trains of action potentials with a rather marked decline of discharge frequency over time (Fig. 1B). The first interval between action potentials had a mean duration of 64.5 + 37.5 ms and the last interval of 159.0 + 89.3 ms (8-10 action potentials/pulse). Following induction of 5 - 6 action potentials during an 100 ms depolarizing current injection all these cells displayed a fast afterhyperpolarization (AHP) that peaked 24.0 + 10.3 ms after the end of the current pulse. The amplitude of the AHP was -3.0 + 1.4 mV (n = 14). Thirteen of these ceils showed also a slow afterdepolarization o f - 1 . 5 + 1.3
mV that was measured at its peak or, if the afterdepolarization displayed a gradual decay, 300 ms after the end of the current injection (Fig. 1C). Upon injections of hyperpolarizing current (100 ms) applied at rest, all regular firing cells displayed an inward rectification ('sag') and a slow afterdepolarization at the end of the current pulses that increased with rising injection of hyperpolarizing current (Fig. 1E). In some neurones the afterdepolarization reached threshold and consequently elicited an action potential, but never a bursting discharge. Fig. 1D shows that both, the 'sag' potential and the slow afterdepolarization decreased, when the MP was brought to more negative levels (15-20 mV below rest). The second group of subicular neurons (n = 46) responded to depolarizing current pulses with a brief burst of 2 - 3 action potentials. These cells showed passive membrane characteristics similar to those of regular firing neurons. The cells of this group had a mean resting membrane potential o f - 6 7 . 0 + 6.8 mV significantly different from regular firing cells (P < 0.05). Input resistance of 41.3 + 15.9 Mfl, the average of the action potential amplitudes of 77.5 + 8.4 mV, the action potential duration of 0.73 + 0.12 ms and mean threshold o f - 5 5 . 2 + 8.0 mV (n = 37) were
C
A
25 mV
25 mV
100ms
500ms 0.3nA
D
E (nA)
AI
-
-77mV~ - 87mV~
FL
1.4nA
2O
- 67rnV
•
5 mV 500ms
1
1
~
0.2 0.4
-20 -10 --
25mY
-30
50 ms -40
- - I
1.0nA
Fig. 1. Electrophysiological properties of subicular regular firing neurons. (A) Response to intracellular current injection at resting membrane potential. Pulse duration, 40 ms; current amplitudes, 0.05-0.3 nA, 0.05 nA increments. (B) Firing pattern and accommodation property near threshold following intracellular current injection. Pulse duration, 800 ms; current amplitude, 0.3 nA. Note the marked decline of discharge frequency over time. (C) Slow and fast afterdepolarization of regular firing cell following induction of 5-6 action potentials during a 100 ms depolarizing current pulse. (D) Voltage responses to intracellularly injected current pulses, voltage dependency of inward rectification and anodal break response. Pulse duration, 100 ms; current amplitudes, -1.0 to -0.2 nA, 0.2 nA steps. Note, (1) Time-dependent inward rectification and depolarizing afterpotential which increase with higher hyperpolarizing current injections, (2) at a membrane potential 15-20 mV more negative than the resting membrane potential inward rectification and anodal break response disappeared. (E) Mean plot of the current-voltage relationship (I-V) measured at the peak (0) and the 'steady state' (&) of the hyperpolarizing and depolarizing voltage deflections and at the peak (B) of the depolarizing afterpotential following 100 ms current pulses.
J. Behr et al. / Neuroscience Letters 220 (1996) 41-44
B
A
C
25 mV
100ms
0.2hA
L- . . . . . 500ms
j
]
E
D -65 mV ~
-75mV .
- 82 mV ~_~
~
25 mV
5my
0.25nA
I 1.5 nA
2O
~
~
43
I {hA)
.
-
~
~
1
~
~
-100.2 0.4
25 mV -,3O -40
Fig. 2. Electrophysiological ploperties of subicular bursting neurons. (A) Response to intracellular current injection at resting membrane potential. Pulse duration, 40 ms; current amplitudes, 0.05-0.2 nA, 0.05 nA increments. Note, (1) Short depolarizing current pulse elicited a self-sustained burst discharge of two action potentials which continued after the end of the pulse, (2) burst is followed by a pronounced hyperpolarizing afterpotential. (B) Firing pattern and accommodation property near threshold following intracellular current injection. Pulse duration, 800 ms; current amplitude, 0.25 nA. Note, Bursting cell displayed little frequency accommodation over time. (C) Afterdepolarization of bursting cell following induction of a burst and 3-4 action potentials during a 100 ms depolarizing current pulse. Note that in contrast to regular firing cells this cell type does not present a slow afterdepolarization. (D) Voltage responses to intracell~alarly injected current pulses, voltage dependency of inward rectification and anodal break response. Pulse duration, 100 ms; current amplitudes, -1.0 to -0.2 nA, 0.2 nA steps. Note, (1) Time-dependent inward rectification and depolarizing afterpotential which increase with higher hyperpolarizing currenl injections, (2) at a membrane potential 15-20 mV more negative than the resting membrane potential inward rectification and anodal break response disappeared. (E) Mean plot of the current-voltage relationship (l-V) measured at the peak (@) and the 'steady state' (A) of the hyperpolarizing and depolarizing voltage deflections and at the peak (D) of the depolarizing afterpotential (anodal break response) following 100 ms current pulses.
not significantly different from regular firing cells (P >
0.05). Application of short depolarizing current pulses (20-40 ms) elicited in contrast to regular firing cells a self-sustained burst discharge of 2 - 3 action potentials which continued after the end of the pulse (Fig. 2A). The burst was followed by a pronounced negative afterpotential of --6.5 + 3.1 mV that peaked 29.3 + 7.2 ms after termination of the depolarizing cun'ent pulse (n = 7). Depolarization above threshold for 800 ; m s (0.05-2.0 nA) caused an initial burst of action potentials followed by an afterhyperpolarization and then by a train of single action potentials with a frequency depending on the magnitude of the depolarization (Fig. 2B). In contrast to regular firing cells bursting cells displayed little frequency accommodation. The interval between the first and second action potential following the burst was 83.6 + 30.6 ms and between the before last and the last action potential 87.7 + 33.6 ms (8-10 action potentials/pulse). Bursting cells displayed an afterhyperpolarization of -3.0 + 1.3 mV that peaked 22.5 + 5.0 ms after the end of a 100 ms depolarizing current pulse that induced a burst and 3 - 4 action potentials. The amplitude
of this fast afterhyperpolarization showed no difference to regular firing cells (P < 0.05). Twelve of 14 investigated bursting neurons did not present a slow afterdepolarization in contrast to regular firing cells (Fig. 2C) (P = 0.001). In response to hyperpolarizing current pulses bursting cells displayed also a 'sag' potential and a slow afterdepolarization that could be sufficient to elicit action potentials or burst discharges. Both could be blocked bY hyperpolarizing background currents that hyperpolarized the cell by 15-20 mV below resting potential (Fig. 2D). The dependence of the 'sag' potential and the slow afterdepolarization on membrane potential was similar to that of the regular firing neurons (Fig. 2E). The present study confirms a high incidence of bursting cells in the subiculum. We could show that beyond their firing pattern these cell type present distinct electrophysiological differences from regular firing cells. Like in previous studies the major division of subicular neurons was into burster and regular firing cells. Bursting could be induced either by depolarizing current pulses or by afterdepolarizations following the end of hyperpolarizing current injections. The burst generating mechanisms
44
J. Behr et al. / Neuroscience Letters 220 (1996) 41-44
underlying hyperpolarization induced bursting may be different from that following depolarization. Hyperpolarization induced bursting was previously shown to be sensitive to tetrodotoxin (TTX) [8] while depolarization induced bursting in the same type of neurons seemed to be insensitive to TTX [13]. The incidence of burster cells was in our study somewhat smaller than reported by Mattia et al. [8], Stewart and Wong [13], Mason [6] and Taube [15]. The incidence of burster cells in those studies was 60100%. As bursting cells can show single-spiking behavior when their membrane is depolarized, we always examined the firing patterns from a membrane potential of about -80 to -70 mV in order to avoid false classification. Differences in cell properties may also result from previous exposure to neuromodulators. Preliminary experiments have revealed that these neurons do not change their discharge behavior after previous exposure to 5-HT (Behr et al., submitted). Alternatively, spontaneous transmitter release may also influence discharge behavior. In our hands the discharge mode of a given neuron remained unaltered, when glutamate and GABA receptors were blocked. The difference in the incidence of burster cells may thus vary between subfields of the subiculum and along the dorso-ventral axis of the hippocampus. Previously reports have shown such variations in area CA1 [7]. Both cell types expressed upon hyperpolarization a delayed inward rectification with a subsequent afterdepolarization as has been previously reported [8,13,15]. The 'sag' potential was previously shown to be blocked by Cs ÷ [8,13] suggesting that it is dependent on an inward rectifying current consisting of a mixed Na÷/K ÷ conductance (IQ or IH) or an K ÷ inward rectifier. Indeed, activation of these currents by constant depolarization let the delayed inward rectification disappear. Besides the common intrinsic characteristics bursting and regular firing cells differ significantly with regard to accommodation properties and afterhyperpolarization. The lack of accommodation in bursting cells might be related to the absence of a slow AHP. The fast repolarization in bursting cells following an action potential facilitates the generation of subsequent action potentials whereas in regular firing cells Ca2÷-dependent slow AHPs that increase during a train of action potentials, might cause the marked accommodation. In contrast, bursting cells of the neocortex may present a slow AHP and during depolarizing current pulses their spike frequency does adapt [9]. Why does the subiculum contain a relatively extensive subpopulation of bursting cells with distinct intrinsic properties? An essential function of bursting cells is the amplification of neural signals [5]. As the subiculum is one of the important output stations for the hippocampal complex, its subpopulation of bursting neurons might serve as an amplifier facilitating the processing of information. This hypothesis may be confirmed by the fact that bursting cells do not display a marked accommodation. Above that
the hippocampal-entorhinal complex is known to be an epilepsy-prone area. It is known that the subiculum participates strongly in ictaform activity [2]. Therefore subicular bursting cells may support the generation and spread of convulsive activity. This work was supported by a grant from a DFG Innovationskolleg INK 21/Al-1 and the BMBF. We are indebted to H. Tetsch and A. Dtierkop for technical assistance in the experiments. [1] Amaral, D.G., Dolorfo, C. and Alvarez-Royo, P., Organization of CA1 projections to the subiculum: a PHA-L analysis in the rat, Hippocampus, 1 (1991)415-436. [2] Behr, J. and Heinemann, U., Low Mg 2÷ induced epilptiform activity in the subiculum before and after disconnection from rat hippocampal and entorhinal cortex slices, Neurosci. Lett., 205 (1996) 25-28. [3] Connors, B.W., Gutnick, M.J. and Prince, D.A., Electrophysiological properties of neocortical neurons in vitro, J. Neurophysiol., 48 (1982) 1302-1320. [4] Finch, D.M. and Babb, T.L., Demons~ation of caudally directed hippocampal efferents in the rat by intracellular injection of horseradish peroxidase, Brain Res., 214 (1981) 405-410. [5] Hablitz, J.J. and Johnston, D., Endogenous nature of spontaneous bursting in hippocampal pyramidal neurons, Cell. Mol. Neurobiol., 1 (1981) 325-333. [6] Mason, A., Electrophysiology and burst-firing of rat subicular pyramidal neurons in vitro: A comparison with area CA1, Brain Res., 600 (1993) 174-178. [7] Masukawa, L.M., Benardo, L.S. and Prince, D.A., Variations in electrophysiological properties of hippocampal neurons in different subfields, Brain Res., 242 (1982) 341-344. [8] Mattia, D., Hwa, G.G. and Avoli, M., Membrane properties of rat subicular neurons in vitro, J. Neurophysiol., 70 (1993) 1244-1248. [9] McCormick, D.A., Connors, B.W., Lighthall, J.W. and Prince, D.A., Comparative electrophysiology of pyramidal and sparsely spiny steUate neurons of the neocortex, J. Neurophysiol., 54 (1985) 782-806. [10] S c h w ~ o i n , P.A., Characteristics of CA1 neurons recorded intracellularly in the hippocampal 'in vitro' slice preparation, Brain Res., 85 (1975) 423-436. [11] Schwartzkroin, P.A., Further characteristics of hippocampal CA1 cells in vitro, Brain Res., 128 (1977) 53-68. [12] Steward, O. and Scoville, S.A., Cells of origin of entorhinal cortical afferents to the hippocampus and fascia dentata of the rat, J. Comp. Neurol., 169 (1976) 347-370. [13] Stewart, M. and Wong, R.K., Intrinsic properties and evoked responses of guinea pig subicular neurons in vitro, J. Neurophysiol., 70 (1993) 232-245. [14] Tamamaki, N. and Nojyo, Y., Disposition of the slab-like modules formed by axon branches originating from single CA1 pyramidal neurons in the rat hippocampus, J. Comp. Neurol., 291 (1990) 509-519. [15] Taube, J.S., Electrophysiological properties of neurons in the rat subiculum in vitro, Exp. Brain Res., 96 (1993) 304-318. [16] Witter, M.P., Organization of the entorhinal-hippocampal system: a review of current anatomical data, Hippocampus, 3 (1993) 33-44. [17] Witter, M.P. and Groenewegen, H.J., The subiculum: cytoarchitectonically a simple structure, but hodologically complex, Prog. Brain Res., 83 (1990) 47-58. [18] Witter, M.P., Ostendorf, R.H. and Groenewegen, H.J., Heterogeneity in the dorsal subiculum of the rat: distinct neuronal zones project to different cortical and subcortical targets, Eur. J. Neurosei., 2 (1990) 718-725.