| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1 Department of Neuroscience, The Johns Hopkins University School of Medicine, 725 N. Wolfe Street, 916 Hunterian Building, Baltimore, MD 21205, USA
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionSupplementary materialReferences |
|---|
(Received 12 July 2004;
accepted after revision 18 October 2004;
first published online 21 October 2004)
Corresponding author D. J. Linden: Department of Neuroscience, The Johns Hopkins University School of Medicine, 725 N. Wolfe Street, 916 Hunterian Building, Baltimore, MD 21205, USA. Email: dlinden{at}jhmi.edu
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionSupplementary materialReferences |
|---|
DCN neurones in vivo typically fire spontaneously at 10–20 Hz (Thach, 1968; McDevitt et al. 1987; LeDoux et al. 1998). Artificial injection of depolarizing current can increase this firing frequency up to 300 Hz (Jahnsen, 1986). DCN neurones also show rebound excitation following a burst of IPSPs or injection of hyperpolarizing current. Repolarization is often accompanied by a ‘rebound’ depolarizing envelope (driven by Ih, a persistent Na+ conductance and low-threshold Ca2+ current) which triggers Na+ spike firing (Jahnsen, 1986; Llinas & Muhlethaler, 1988; Aizenman & Linden, 1999; Raman et al. 2000). The rebound depolarization is terminated in part by the opening of SK-type Ca2+-sensitive K+ channels (Aizenman & Linden, 1999). Thus, DCN neurones can respond to either the onset of EPSP bursts or the offset of IPSP bursts with transient increases in firing rate.
Several lines of evidence have implicated the DCN in the storage of certain forms of motor learning, particularly associative eyelid conditioning. Extracellular recordings made from behaving rabbits have shown that DCN neurones (particularly those in the interposed nucleus) reflect the memory trace for associative eyelid conditioning. As animals acquire the shock–tone association, DCN neurones begin to fire strongly immediately prior to shock onset. Furthermore, lesions, inactivation and local blockade of protein synthesis in the DCN can prevent acquisition of this task when applied before training and can eliminate the memory for training when applied afterwards (see Lavond (2002) for review). One cellular explanation for this memory trace has been that associative training ultimately produces long-term potentiation of those mossy fibre–DCN synapses that convey tone information (Medina et al. 2000). Another possible memory mechanism could involve changes in the intrinsic excitability of DCN neurones produced by training (see Hansel et al. (2001) for review). In this scheme, persistent changes in the properties of voltage-gated ion channels could result in altered firing rate or firing pattern as well as changes in the active, integrative properties of dendrites such as synaptic summation and spike backpropagation.
Indeed, there are now many examples, in both vertebrate and invertebrate model systems, of rapid, persistent changes in intrinsic neuronal excitability produced by behavioural training in intact animals or artificial synaptic stimulation in reduced preparations such as brain slices or cell cultures (see Zhang & Linden (2003) for review). In the DCN, prior work has shown that application of EPSP bursts to DCN neurones in a rat brain slice preparation gives rise to an increase in intrinsic excitability. Test pulses consisting of a fixed injection of depolarizing current evoke a greater number of spikes after a conditioning stimulus composed of EPSP bursts, and this increase persists for as long as the recordings can be maintained (>30 min). This phenomenon was blocked by application of an NMDA receptor antagonist and could be mimicked by repeated large injections of depolarizing current (Aizenman & Linden, 2000).
Here, we have combined microelectrode recording, whole-cell current-clamp recording and Ca2+ imaging to characterize intrinsic plasticity in the DCN of juvenile rat brain slices. We have sought to address the following questions. What electrophysiological sequelae accompany the induction in intrinsic excitability increases? Can increases in intrinsic excitability be manifested as changes in firing pattern? What is the range of stimuli that can induce increases in intrinsic excitability?
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionSupplementary materialReferences |
|---|
Cerebellar slices were prepared as previously described (Aizenman et al. 1998). Briefly, juvenile (12- to 15-day-old) Sprague Dawley rats were killed by decapitation and the cerebellum was quickly removed and cooled with ice-cold standard artificial cerebrospinal fluid (ACSF). This procedure was approved by the Animal Care and Use Committee of The Johns Hopkins University School of Medicine. ACSF contained (mM): 124 NaCl, 5 KCl, 2.5 CaCl2, 1.5 MgSO4, 26 NaHCO3, 1.25 NaH2PO4, and 20 D-glucose, and was equilibrated with 95% O2–5% CO2 to yield pH 7.4. Coronal slices of the cerebellum (400 µm thick) were cut on a vibrating tissue slicer (Leica VT1000S). After a 1 h recovery period at room temperature, slices were transferred to a Haas-style interface chamber. During recording, slices were continuously perfused at 32°C with standard ACSF at 2 ml min–1. In some experiments, 200–300 µM picrotoxin or 2 mM kynurenic acid were added to the ACSF to block GABAA receptors and ionotropic glutamate receptors, respectively. Microelectrode intracellular recordings were made with borosilicate glass electrodes (100–150 M
) filled with 3 M K-acetate. A small hyperpolarizing bias current was applied to hold the membrane potential just below spike threshold. Synaptic tetani were applied via concentric bipolar stimulating electrodes placed in the white matter adjacent to the nuclei. Recordings were made from either the medial or lateral group of the DCN, and no differences were observed between the two nuclei. Excitatory synaptic stimuli (EPSP bursts), consisted of 10 bursts applied at 4 Hz; each burst comprising 10 pulses delivered at 100 Hz, repeated five times. EPSP bursts were delivered with 200–300 µM picrotoxin in the ACSF to block GABAA-ergic transmission. Inhibitory synaptic stimuli (IPSP bursts) consisted of 20 bursts applied at 1 Hz, each burst comprising 10 pulses delivered at 100 Hz, applied with 2 mM kynurenic acid in the ACSF to block ionotropic glutamatergic transmission. This stimulation regime evoked a large number of rebound spikes. To record changes in intrinsic excitability, short depolarizing test pulses (300 ms, 0.05–0.3 nA) were applied. Recordings were made using an Axoclamp-2A amplifier (Axon Instruments). Recordings were filtered at 10 kHz, digitized at 10 kHz, and collected with a Macintosh computer running AxoGraph 4.6 software (Axon Instruments). Kynurenic acid was purchased from Tocris, UK. All other drugs were obtained from Sigma.
Data analysis
Data were analysed using AxoGraph 4.6 and Igor Pro software (WaveMetrics). The threshold for the first evoked spike (see Fig. 1D) was measured at the first point where the slope reached 20 V s–1. The spike latency was measured from the onset of the test pulse to the peak of the first evoked spike. The slope of the depolarizing prepotential was calculated using a window 15–10 ms before the first evoked spike threshold. To measure the input resistance (Rinput), a 100 ms-long, – 0.1 nA test pulse was delivered. Rinput was calculated by curve fitting to the equation:
|
|
|
|
|
|
Whole-cell patch recording was performed using 250-µm-thick coronal cerebellar slices. After a 1 h recovery period, slices were placed in a submerged chamber superfused with ACSF at 32°C. DCN neurones in the medial or lateral group were visualized through a 40 x water immersion objective using infrared Dodt gradient contrast optics (Luigs & Neumann, Ratingen, Germany), and whole-cell patch-clamp recordings were made with electrodes (3–6 M) filled with internal saline containing (mM): 135 K-gluconate, 10 KCl, 10 Hepes, 4 Na2-ATP, 0.4 Na-GTP, and 0.2 Oregon Green BAPTA-1, pH 7.2, osmolarity 290 for calcium imaging. To measure neuronal excitability after EPSP bursts, electrodes were filled with internal saline containing (mM): 135 KCH3SO3, 5 KCl, 10 Hepes, 0.2 EGTA, 4 Na2-ATP, 2 MgCl2, 0.4 Na3-GTP for control neurones, and filled with internal saline containing (mM): 111 KCH3SO3, 3 KCl, 10 Hepes, 12 BAPTA-K4, 1 MgCl2, 2 CaCl2, 4 Na2-ATP, 0.4 Na3-GTP for BAPTA neurones. The ACSF was supplemented with 1 µM strychnine and 300 µM picrotoxin. To measure neuronal excitability after IPSP bursts, electrodes were filled with internal saline containing (mM): 125 KCH3SO3, 5 KCl, 10 Hepes, 4 Na2-ATP, 0.4 Na3-GTP, 2 MgCl2, 0.2 EGTA, 10 phosphocreatine (di-tris salt) for control neurones, and filled with internal saline containing (mM): 98 KCH3SO3, 3 KCl, 10 Hepes, 4 Na2-ATP, 0.4 Na3-GTP, 1 MgCl2, 12 BAPTA-K4, 2 CaCl2, 10 phosphocreatine (di-tris salt) for BAPTA neurones. The ACSF was supplemented with 3 mM kynurenic acid. Series resistance was <25 M and was compensated at 70–80%. Synaptic burst stimulation was applied via a monopolar patch electrode filled with ACSF and placed in the white matter adjacent to the nuclei. Recordings were performed in current-clamp mode with either an Axopatch-1C or Axopatch 200B amplifier (Axon Instruments). Recordings of membrane voltage were filtered at 5 kHz, digitized at 10 kHz and collected with pClamp 9 software (Axon instruments). The imaging experiments began at least 30 min after break-in to allow for dye diffusion. Ca2+ imaging was performed with a Zeiss LSM 510 confocal microscope, using the 488 nm line of an argon laser for excitation, and emitted green fluorescence was detected through a 505 nm long-pass filter. The dendrites of DCN neurones typically elaborate in a three-dimensional fashion making it difficult to capture distal dendrites, proximal dendrites and soma in a single focal plane. To allow for simultaneous measurement in these compartments the pinhole was set wide open to maximize depth of field. A 128 x 128 pixel frame (1.3–2.5 µm pixel–1) was scanned at a rate of 5–10 Hz. Background correction was performed by subtracting the background fluorescence of a representative neighbouring region next to the soma, proximal dendrite, or distal dendrite. Ca2+ signal amplitudes were expressed as (Ft
–
F0)/F0. The average fluorescence intensity in the baseline period was taken as F0. At the end of experiment, the pinhole was closed to yield an Airy value of 1, z-stack confocal images were acquired and the neurone's image was reconstructed. Images were collected with Zeiss LSM software and analysed with Zeiss LSM and Igor Pro software. Oregon Green BAPTA-1 was purchased from Molecular Probes.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionSupplementary materialReferences |
|---|
Intrinsic excitability was measured by continually injecting a small hyperpolarizing bias current to silence ongoing spike firing, and then injecting a 300-ms-long depolarizing current (0.1–0.2 nA) sufficient to evoke a small number of spikes (typically 2–3; Fig. 1A). These experiments were performed in picrotoxin-containing external saline to block GABAA receptors. The depolarizing test pulse was delivered every 15 s and responses were fairly stable over a 25 min recording period (Figs 1B and C; mean change in number of spikes: 0.55 ± 0.25 spikes; 22 ± 12% change relative to baseline; n = 9). In a separate population of cells, after a 5-min period of baseline recording, a conditioning stimulus was given consisting of high-frequency burst stimuli delivered through a stimulating electrode placed in the adjacent white matter to evoke EPSPs. This resulted in a slow, sustained increase in the number of action potentials evoked by the depolarizing test pulse (Fig. 1B and C; 2.82 ± 0.45 spikes; 152 ± 53% change; measured at t = 20–25 min; n = 11, P < 0.05 compared to control group) as previously reported (Aizenman & Linden, 2000). A number of other measurements could also be derived from this experiment. EPSP bursts resulted in a significant reduction of spike threshold (Fig. 1E; measured for the first evoked spike; –3.9 ± 0.5 mV) and the latency to the first spike (–53 ± 4% change). The slope of the depolarizing prepotential (which intervenes between current injection and the first spike threshold) was also significantly increased following EPSP bursts (67 ± 11% change). Rinput, monitored using a 100-ms-long, 0.1 nA hyperpolarizing test pulse, was also persistently increased by the conditioning stimulation, although this was a small effect (19 ± 5% change). The increase in Rinput is unlikely to underlie the persistent increase in the number of evoked spikes as these two measures were uncorrelated across the population of cells which received EPSP bursts (r2 = 0.067, P > 0.05; Supplementary Fig. 1). Overall, the persistent effects of EPSP bursts may reflect an alteration of one or more types of voltage-gated ion channels, some of which operate at or below spike threshold.
One current which has an important role in regulating subthreshold excitability is the hyperpolarization-activated cation conductance, Ih. This current is subject to both short- and long-term modulation. Previous work has shown that induction of febrile seizures in young rats results in a persistent increase of Ih in hippocampal pyramidal neurones (Chen et al. 2001). Application of the anticonvulsant drug limotrigine also decreases dendritic excitability through an increase in dendritic Ih in hippocampal pyramidal neurones (Poolos et al. 2002). To test the hypothesis that persistent increases in intrinsic excitability in the DCN are associated with alterations in Ih, we used 300-ms-long injections of negative current as test pulses (–0.3 to –0.5 nA). These test pulses evoked an initial hyperpolarization that was then attenuated by the development of a depolarizing ‘sag’ in the voltage response, typically attributed to Ih. The amplitude of this depolarizing sag is greater with larger injections of negative current (Jafri & Weinreich, 1998; Aizenman & Linden, 1999; Chen et al. 2001; Williams et al. 2002). Following the termination of the test pulse there is a rebound depolarization, the size of which is also proportional to the peak amplitude of the hyperpolarization (Aizenman & Linden, 1999). If sufficiently large, this rebound depolarization can evoke one or more Na+ spikes. Following, EPSP bursts, the injection of a constant test pulse gave rise to a larger peak hyperpolarization, which can be attributed to the previously measured increase in Rinput (Fig. 1E). Consistent with previous reports (Jafri & Weinreich, 1998), this caused a greater degree of sag as a consequence of the larger peak hyperpolarization. However, when the amplitude of the negative current test pulse was reduced to match the baseline peak hyperpolarization, the amplitude of the depolarizing sag was unchanged (Fig. 2A and C; 5.1 ± 0.8 mV before EPSP bursts, 5.3 ± 0.8 mV after EPSP bursts, n = 5). Therefore, modulation of Ih is unlikely to underlie persistent increases in intrinsic excitability produced by EPSP bursts. Interestingly, the probability of evoking an action potential during the subsequent rebound depolarization was increased, even when the test pulses were reduced after EPSP bursts to match the baseline peak hyperpolarization (Fig. 2A and B; P = 0.055).
Bursts of Na+ spikes are often followed by an afterpotential, either an afterhyperpolarization (AHP) or an afterdepolarization (ADP). Several conductances can contribute to these afterpotentials including the Ca2+- and voltage-sensitive current, IC, voltage-sensitive Ca2+ currents, the Ca2+-sensitive K+-current, mIAHP, the voltage-sensitive K+-current, IM, and the Ca2+-sensitive K+-current, sIAHP (see Storm (1990) for review). To determine whether these afterpotentials are persistently altered following EPSP bursts we used a hybrid stimulation protocol. First, responses to the standard 300-ms-long depolarizing test pulse were recorded for 5 min and then we switched to a compound test pulse consisting of four 2-ms-long large positive current injections (2 nA) delivered every 7 ms repeated every 60 s. This compound test pulse evoked a set of four Na+ spikes with consistent timing, followed by an afterpotential. After delivery of the EPSP burst conditioning stimulus, simple depolarizing step test pulses were resumed for 20 min, followed by another set of compound test pulses (Fig. 3). The main finding of this experiment was that EPSP bursts produced a significant shift in the afterpotential in the hyperpolarizing direction (Fig. 3B and C; –1.5 ± 0.2 mV, n = 5), sometimes resulting in the conversion of an ADP to an AHP. This increase in the rate and extent of repolarization may help to speed Na+ channel de-inactivation thereby supporting an increase in spike frequency.
Consistent with previous findings (Aizenman & Linden, 2000; Fig. 1), the standard 300-ms-long test pulses revealed a significant increase in the number of evoked spikes following EPSP bursts (Fig. 3A; 2.8 ± 0.3 spikes, 109 ± 20% increase, n = 5). However, the control group also showed a small increase in excitability (0.6 ± 0.2 spikes, 38 ± 20% increase, n = 5), probably because the compound test pulses themselves were sufficient to evoke a small degree of intrinsic plasticity. Despite the small increase in excitability of the control group, neurones in the EPSP burst group still showed a significant increase in the number of evoked spikes when compared to control (P < 0.05).
To this point, we have only considered intrinsic plasticity in tonic firing DCN neurones. Bursting neurones show a range of firing patterns in response to a 300 ms depolarizing step. Some show a simple burst-and relax-pattern (Fig. 4A), while others may fire one or two spikes at lower frequency before burst onset. The duration of the burst can also vary considerably from neurone to neurone. However, for a single neurone, the pattern of spiking in response to a depolarizing test pulse is reasonably stable (Fig. 4A). Following EPSP bursts, intrinsically burst-firing cells showed one of two forms of modulation: 6 of 12 cells showed a persistent increase in the number of depolarization-evoked spikes similar to that seen in tonic firing cells (Fig. 4B; 82 ± 11% change, P < 0.01 compared to control group). The other six cells showed no significant change in the number of spikes (–9 ± 6% change, P = 0.31 compared to control group), but rather a shift in the firing mode from bursting to tonic firing. This was reflected in a significant increase of the first interspike interval in the cells which showed a firing mode shift (Fig. 4D; 381 ± 111% change) but not in those in which the number of evoked spikes was increased (6 ± 13% change), nor in control cells (8 ± 8% change, n = 7). To further quantify the firing mode, interspike intervals were calculated and normalized to their mean value. The S.D. of the normalized interspike intervals was then calculated. In the control group, this value was quite stable over the recording period (Fig. 4C; –2 ± 3% change). However, in the cells which received EPSP bursts and showed a mode-shift, the percentage change in the S.D. of the normalized interspike interval decreased (–32 ± 9% change, P < 0.01 compared to control group), indicating a shift to a more tonic firing pattern.
What accounts for the difference between the bursting cells which show an increase in the number of evoked spikes versus those that show no increase but shift firing mode? These two groups showed similar reductions in the latency of the first evoked spike and a similar small increase in Rinput. One difference emerged when examining a baseline input/output function for depolarizing current injection. Post hoc analysis showed that those cells which responded to EPSP bursts with a firing mode-shift were less excitable in the baseline state: they produced fewer spikes in response to increasing 300-ms-long injections of depolarizing current (Fig. 4E).
DCN neurones show rebound depolarization after stimulation by IPSP bursts. Indeed, previous work has shown that IPSP bursts can trigger LTP and LTD of the Purkinje cell–DCN synapse (Aizenman et al. 1998). We investigated whether IPSP bursts could similarly trigger changes in intrinsic excitability in tonic firing DCN neurones. IPSP bursts (each consisting of 10 pulses at 100 Hz, repeated with a burst frequency of 1 Hz for 20 repetitions) caused DCN neurones to respond to depolarizing test pulses with a larger number of spikes (Fig. 5A–C; 3.2 ± 0.3 spikes, 169 ± 26% change, n = 6, P < 0.01 compared to control group). As seen previously with EPSP bursts, this was associated with a reduction in spike threshold (Fig. 5D; –2.2 ± 0.2 mV) and first spike latency (–58 ± 3% change) as well an increase in the slope of the depolarizing prepotential (77 ± 17% change) and Rinput. (9 ± 2% change). The increase in Rinput is unlikely to underlie the persistent increase in the number of evoked spikes as these two measures were uncorrelated across the population of cells which received IPSP bursts (r2 = 0.0003, P > 0.05; Supplementary Fig. 1). Hence, at least in tonic firing DCN neurones, IPSP and EPSP bursts can produce similar persistent increases in intrinsic excitability.
|
Whole-cell current-clamp recording was used to measure membrane voltage and to deliver the Ca2+ reporter dye Oregon Green BAPTA-1, which was excited by an argon ion laser on a scanning confocal microscope. Following a period of 
30 min for dye equilibration, measurements were made using the microscope with the pinhole wide open in order to maximize depth of field and follow the course of a dendrite through the z-dimension, and thereby allow for simultaneous measurement of multiple regions (Fig. 6B). The proximal dendrite was defined as that within 50 µm (three-dimensional distance) of the soma while the distal dendrite was at least 100 µm (three-dimensional distance) from the soma. EPSP bursts (identical to those used previously for intrinsic plasticity experiments) produced bursts of spikes. A set of 10 bursts of 10 EPSPs each produced a Ca2+ transient which was largest in the distal dendrite (Fig. 6A; 
F/F
= 0.68 for the peak transient of the first set) and which was diminished in the proximal dendrite (F/F
= 0.31) and further reduced in the soma (F/F
= 0.07). This pattern was also reflected in the population of cells measured in this fashion: the first peak amplitude of the Ca2+ response in the proximal dendrite was 45%
± 6% of that in the distal dendrite (n
= 4).
|
F/F
= 0.49 at high gain) and decreased in the proximal dendrite (F/F
= 0.21 at high gain, 0.31 at low gain) and further decreased in the soma (F/F
= 0.12 at low gain). The first peak amplitude of the Ca2+ response in the proximal dendrite was 54%
± 11% of that in the distal dendrite (n
= 3). The distal dendritic response showed individual peaks evoked by each burst superimposed upon a plateau, whereas the individual burst responses in the proximal dendrite and soma decayed so slowly as to fuse. Thus, both EPSP and IPSP bursts can drive Ca2+ responses which are larger in amplitude in the distal dendrites compared to the proximal dendrites and soma.
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionSupplementary materialReferences |
|---|
Intrinsically bursting DCN neurones respond to EPSP bursts with one of two types of persistent change: either an increase in excitability manifested as an increase in evoked spike frequency, or a conversion from a bursting to a tonic firing pattern (Fig. 4A). Why do some intrinsically bursting DCN cells respond to EPSP bursts with a mode shift to tonic firing while others show a different response consisting of increasing the number of evoked spikes while retaining their previous burst-firing mode? One clue comes from a retrospective analysis showing that, in the basal state, mode-shifting cells responded to large positive current injections with smaller increases in firing rate (Fig. 4E). Another comes from the observation that the number of spikes evoked by EPSP bursts during the conditioning stimulation was similar between these two groups (data not shown). We suggest that mode-shifting and spike-increasing responses to EPSP bursts reflect a similar alteration of intrinsic conductances, merely superimposed upon two different baseline excitability profiles. A stringent test of this hypothesis will ultimately require identification of these conductances and their measurement with voltage-clamp recordings before and after EPSP bursts. K+ and Ca2+-sensitive K+ conductances are particularly good candidates for mediating both excitability increases and mode-shifting.
While rapid, persistent synaptically driven changes in firing mode had not been previously reported, there is precedent for slower modulation of this parameter. In the stomatogastric ganglion of the spiny lobster Panulirus or the crab Cancer, identified neurones function in a circuit that generates gastric mill rhythms (Harris-Warrick, 2002). In vivo, these neurones are exposed to synaptic and neuromodulatory drive and they fire in bursts. When deprived of this input soon after being placed in culture, these neurones are mostly silent and respond to depolarizing current injection with tonic firing. However, after 2–4 days in culture (while still isolated from synaptic input), these neurones changed their activity from tonic firing to burst firing, thereby re-acquiring aspects of their in vivo firing pattern (Turrigiano et al. 1994). This involved upregulation of several inward currents including voltage-sensitive Ca2+ current, a rapidly inactivating voltage-sensitive Na+ current and a slowly inactivating Na+ plateau current, and downregulation of the transient K+ current IA and a delayed outward rectifier K+ current (Turrigiano et al. 1995). At present, it is unclear whether the mechanisms engaged by this slow form of intrinsic plasticity will also be found in rapid changes in firing mode induced by EPSP bursts.
Ca2+ imaging experiments performed herein show that both EPSP bursts and IPSP bursts (and their consequent rebound spiking) produce large Ca2+ transients in the distal dendrite but these transients become smaller and slightly more prolonged (and hence subject to summation) in the soma (Figs 6 and 7). When these IPSP bursts were delivered to tonic firing DCN neurones, they produced essentially the same effect as EPSP bursts: an increase in average evoked firing rate and the set of changes which accompany this (reduced spike threshold, etc.). There is also precedent for driving persistent increases in intrinsic excitability by activating inhibitory synapses. In mouse brainstem slices, a 5 min long periodic stimulation of GABAergic drive to neurones of the medial vestibular nucleus produced long-lasting increases in both the spontaneous firing rate and the increase in firing rate evoked by depolarizing current injection (Nelson et al. 2003). This phenomenon was associated with an increase in Rinput and a decrease in the AHP following single spikes, but not with a change in spike threshold. It was mimicked and occluded by application of iberiotoxin, a specific blocker of large-conductance Ca2+-dependent K+ (BK) channels. Interestingly, this form of intrinsic excitability was not driven by rebound spiking and dendritic Ca2+ transients, as in the present case. Rather, it appeared to result from a reduction in basal Ca2+ concentration during the conditioning stimulus. Thus, there are at least two distinct mechanisms by which repeated activation of inhibitory GABAergic synapses can drive persistent increases in postsynaptic intrinsic excitability.
The observation that IPSP bursts followed by a brief pause are an ideal stimulus to evoke rebound spiking, associated dendritic Ca2+ transients, and ultimately, intrinsic plasticity of DCN neurones has potential implications for cerebellar circuit function. The inferior olive typically signals motor errors through spike firing that is conveyed to cerebellar Purkinje cells through excitatory climbing fibre input. Climbing fibre activation results in a transient acceleration in Purkinje cell firing rate followed by a brief pause (Armstrong et al. 1973; Eccles et al. 1974; Ruigrok, 1997). This will be conveyed via Purkinje cell axons to DCN neurones as an IPSP burst followed by a brief pause. Therefore, repeated activation of the inferior olive (as one might encounter in a motor learning task) might be able to persistently alter DCN intrinsic excitability through a climbing fibre–Purkinje cell disynaptic loop.
There are several caveats in interpreting the present experiments. We have referred to the stimulation delivered to the DCN by stimulating electrodes placed in the adjacent white matter in the presence of the GABAA antagonist picrotoxin as ‘EPSP bursts’. However, the DCN also receive a number of extrinsic modulatory fibres including those utilizing the neurotransmitters acetylcholine (Woolf & Butcher, 1989), serotonin (Chan-Palay, 1976; Takeuchi et al. 1982; Kitzman & Bishop, 1994), noradrenaline (norepinephrine) (Olson & Fuxe, 1971; Eller & Chan-Palay, 1976) and histamine (Panula et al. 1989; Schwartz et al. 1991). DCN interneurones may also release glycine (Chen & Hillman, 1993; Rampon et al. 1996; Baurle & Grusser-Cornehls, 1997). GABAB receptors are unlikely to be activated by synaptic stimulation as previous work has shown GABAB antagonists to have no effects on evoked synaptic currents in DCN (Morishita & Sastry, 1995; Mouginot & Gähwiler, 1995). It is possible that the effects of ‘EPSP bursts’ on intrinsic plasticity may require the actions of neuromodulators. Similarly, ‘IPSP bursts’ in the presence of the ionotropic glutamate receptor antagonist kynurenate may also involve the release of modulatory neurotransmitters. In addition, the postsynaptic group I metabotropic glutamate receptors mGluR1 and mGluR5 are present in the DCN (Fotuhi et al. 1993; Romano et al. 1995). Since both our ‘EPSP burst’ and ‘IPSP burst’ protocols leave metabotropic glutamate receptors unblocked, it is possible that their activation also contributes to DCN intrinsic plasticity.
There is also a possibility that the persistent increases in intrinsic excitability produced by IPSP bursts reflect a persistent attenuation of tonic GABA release. Previous work has shown that under certain conditions IPSP bursts can give rise to LTD of evoked IPSPs recorded in the DCN (Aizenman et al. 1998). For this phenomenon to contribute to intrinsic plasticity, IPSP bursts would have to result in a sustained decrease of tonic GABAergic drive as well. While this is formally possible, we believe it is unlikely, as a similar form of intrinsic plasticity may be triggered by EPSP bursts in conditions where GABAA receptors are blocked.
A final caveat concerns the measurement of Ca2+ transients. We used the high-affinity dye Oregon Green BAPTA-1 (Kd in vitro = 170 nM) at a concentration of 200 µM. This is likely to introduce a high degree of Ca2+ chelation relative to the endogenous buffering capacity of the DCN neurone cytoplasm. In so doing, we may have produced a systematic slowing of both the rising and decay phases of the recorded evoked Ca2+ transients (Sabatini & Regehr, 1998). However, this would not be expected to differentially alter IPSP versus EPSP burst-evoked Ca2+ measurements or somatic versus dendritic measurements.
The present work extends our previous investigation into the modulation of intrinsic excitability of DCN neurones in two important aspects. Firstly, the modulation of the firing pattern of DCN neurones by EPSP bursts is a novel form of plasticity in this neurone. This phenomenon is particularly interesting from a computational point of view in that it might allow for plasticity in the neural representation of information by a spike timing code. Secondly, the discovery that physiologically pertinent patterns of IPSP bursts can cause a paradoxical increase in the excitability of DCN neurones suggests that the climbing fibre–Purkinje cell disynaptic loop may persistently alter DCN excitability. Further work to understand the signal transduction pathways involved in and the ion channels modulated during intrinsic excitability changes will be invaluable in understanding the role of such changes in phenomena such as information storage and signal integration.
|
Supplementary material |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionSupplementary materialReferences |
|---|
Supplementary Figure 1. Increases in intrinsic excitability evoked by EPSP or IPSP bursts are uncorrelated with increases in input resistance.
Supplementary Figure 2. Blockade of excitability increase by BAPTA is not due to saturation of neuronal excitability.
This material can also be found at:
http://www.blackwellpublishing.com/products/journals/suppmat/tjp/tjp602/tjp602sm.htm
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionSupplementary materialReferences |
|---|
Aizenman CD, Huang EJ, Manis PB & Linden DJ (2000). Use-dependent changes in synaptic strength at the Purkinje cell to deep nuclear synapse. Prog Brain Res 124, 257–273.[Medline]
Aizenman
CD
&
Linden
DJ (1999). Regulation of the rebound depolarization and spontaneous firing patterns of deep nuclear neurons in slices of rat cerebellum. J Neurophysiol
82, 1697–1709.
Aizenman CD & Linden DJ (2000). Rapid, synaptically driven increases in the intrinsic excitability of cerebellar deep nuclear neurons. Nature Neurosci 3, 109–111.[CrossRef][Medline]
Aizenman CD, Manis PB & Linden DJ (1998). Polarity of long-term synaptic gain change is related to postsynaptic spike firing at a cerebellar inhibitory synapse. Neuron 21, 827–835.[CrossRef][Medline]
Armano
S, Rossi
P, Taglietti
V
&
D'Angelo
E (2000). Long-term potentiation of intrinsic excitability at the mossy fiber-granule cell synapse of rat cerebellum. J Neurosci
20, 5208–5216.
Armstrong DM, Cogdell B & Harvey RJ (1973). Response of interpositus neurones to nerve stimulation in chloralose anesthetized cats. Brain Res 55, 461–466.[CrossRef][Medline]
Baurle J & Grusser-Cornehls U (1997). Differential number of glycine- and GABA-immunopositive neurons and terminals in the deep cerebellar nuclei of normal and Purkinje cell degeneration mutant mice. J Comp Neurol 382, 443–458.[CrossRef][Medline]
Chan-Palay V (1976). Serotonin axons in the supra- and subependymal plexuses and in the leptomeninges; their roles in local alterations of cerebrospinal fluid and vasomotor activity. Brain Res 102, 103–130.[CrossRef][Medline]
Chan-Palay V (1977). Cerebellar Dentate Nucleus. Springer-Verlag, Berlin.
Chen K, Aradi I, Thon N, Eghbal-Ahmadi M, Baram TZ & Soltesz I (2001). Persistently modified h-channels after complex febrile seizures convert the seizure-induced enhancement of inhibition to hyperexcitability. Nature Med 7, 331–337.[CrossRef][Medline]
Chen S & Hillman DE (1993). Colocalization of neurotransmitters in the deep cerebellar nuclei. J Neurocytol 22, 81–91.[CrossRef][Medline]
Eccles JC, Sabah NH & Taborikova H (1974). Excitatory and inhibitory responses of neurones of the cerebellar fastigial nucleus. Exp Brain Res 19, 61–77.[Medline]
Eller T & Chan-Palay V (1976). Afferents to the cerebellar lateral nucleus. Evidence from retrograde transport of horseradish peroxidase after pressure injections through micropipettes. J Comp Neurol 166, 285–301.[CrossRef][Medline]
Fotuhi M, Sharp AH, Glatt CE, Hwang PM, von Krosigk M, Snyder SH & Dawson TM (1993). Differential localization of phosphoinositide-linked metabotropic glutamate receptor (mGluR1) and the inositol 1,4,5-trisphosphate receptor in rat brain. J Neurosci 13, 2001–2012.[Abstract]
Gauck
V, Thomann
M, Jaeger
D
&
Borst
A (2001). Spatial distribution of low- and high-voltage-activated calcium currents in neurons of the deep cerebellar nuclei. J Neurosci
21, RC158.
Hansel C, Linden DJ & D'Angelo E (2001). Beyond parallel fiber LTD: the diversity of synaptic and non-synaptic plasticity in the cerebellum. Nature Neuroscience 4, 467–475.[Medline]
Harris-Warrick RM (2002). Voltage-sensitive ion channels in rhythmic motor systems. Curr Opin Neurobiol 12, 646–651.[CrossRef][Medline]
Jafri
MS
&
Weinreich
D (1998). Substance P regulates Ih via a NK-1 receptor in vagal sensory neurons of the ferret. J Neurophysiol
79, 769–777.
Jahnsen
H (1986). Electrophysiological characteristics of neurones in the guinea-pig deep cerebellar nuclei in vitro. J Physiol
372, 129–147.
Kitzman PH & Bishop GA (1994). The origin of serotoninergic afferents to the cat's cerebellar nuclei. J Comp Neurol 340, 541–550.[CrossRef][Medline]
Lavond
DG (2002). Role of the nuclei in eyeblink conditioning. Ann N Y Acad Sci
978, 93–105.
LeDoux MS, Hurst DC & Lorden JF (1998). Single-unit activity of cerebellar nuclear cells in the awake genetically dystonic rat. Neuroscience 86, 533–545.[CrossRef][Medline]
Llinas
R
&
Muhlethaler
M (1988). Electrophysiology of guinea-pig cerebellar nuclear cells in the in vitro brain stem-cerebellar preparation. J Physiol
404, 241–258.
McDevitt CJ, EbnerTJ & Bloedel JR (1987). Relationships between simultaneously recorded Purkinje cells and nuclear neurons. Brain Res 425, 1–13.[CrossRef][Medline]
Medina JF, Nores WL, Ohyama T & Mauk MD (2000). Mechanisms of cerebellar learning suggested by eyelid conditioning. Curr Opin Neurobiol 10, 717–724.[CrossRef][Medline]
Morishita W & Sastry BR (1995). Pharmacological characterization of pre- and postsynaptic GABAB receptors in the deep nuclei of rat cerebellar slices. Neuroscience 68, 1127–1137.[CrossRef][Medline]
Mouginot D & Gähwiler BH (1995). Characterization of synaptic connections between cortex and deep nuclei of the rat cerebellum in-vitro. Neuroscience 64, 699–712.[CrossRef][Medline]
Muri
R
&
Knopfel
T (1994). Activity induced elevations of intracellular calcium concentration in neurons of the deep cerebellar nuclei. J Neurophysiol
71, 420–428.
Nelson AB, Krispel CM, Sekirnjak C & du Lac S (2003). Long-lasting increases in intrinsic excitability triggered by inhibition. Neuron 40, 609–620.[CrossRef][Medline]
Olson L & Fuxe K (1971). On the projections from the locus coeruleus noradrealine neurons: the cerebellar innervation. Brain Res 28, 165–171.[CrossRef][Medline]
Panula P, Pirvola U, Auvinen S & Airaksinen MS (1989). Histamine-immunoreactive nerve fibers in the rat brain. Neuroscience 28, 585–610.[CrossRef][Medline]
Poolos NP, Migliore M & Johnston D (2002). Pharmacological upregulation of h-channels reduces the excitability of pyramidal neuron dendrites. Nat Neurosci 5, 767–774.[Medline]
Raman
IM, Gustafson
AE
&
Padgett
D (2000). Ionic currents and spontaneous firing in neurons isolated from the cerebellar nuclei. J Neurosci
20, 9004–9016.
Rampon C, Luppi PH, Fort P, Peyron C & Jouvet M (1996). Distribution of glycine-immunoreactive cell bodies and fibers in the rat brain. Neuroscience 75, 737–755.[CrossRef][Medline]
Romano C, Sesma MA, McDonald CT, O'Malley K, Van den Pol AN & Olney JW (1995). Distribution of metabotropic glutamate receptor mGluR5 immunoreactivity in rat brain. J Comp Neurol 355, 455–469.[CrossRef][Medline]
Ruigrok TJ (1997). Cerebellar nuclei: the olivary connection. Prog Brain Res 114, 167–192.[Medline]
Sabatini
BL
&
Regehr
WG (1998). Optical measurement of presynaptic calcium currents. Biophys J
74, 1549–1563.
Schwartz
JC, Arrang
JM, Garbarg
M, Pollard
H
&
Ruat
M (1991). Histaminergic transmission in the mammalian brain. Physiol Rev
71, 1–51.
Sourdet
V, Russier
M, Daoudal
G, Ankri
N
&
Debanne
D (2003). Long-term enhancement of neuronal excitability and temporal fidelity mediated by metabotropic glutamate receptor subtype 5. J Neurosci
23, 10238–10248.
Storm JF (1990). Potassium currents in hippocampal pyramidal cells. Prog Brain Res 83, 161–187.[Medline]
Takeuchi Y, Kimura H & Sano Y (1982). Immunohistochemical demonstration of serotonin-containing nerve fibers in the cerebellum. Cell Tissue Res 226, 1–12.[Medline]
Thach
WT (1968). Discharge of Purkinje and cerebellar nuclear neurons during rapidly alternating arm movements in the monkey. J Neurophysiol
31, 785–797.
) and EPSP bursts (•) neurones. In EPSP burst experiments, following a 5 min baseline period, the neurone received excitatory synaptic stimuli consisting of 10 bursts applied at 4 Hz, each burst comprising 10 pulses delivered at 100 Hz, repeated five times (indicated by arrowhead). C, population means of percentage change in the number of spikes in control (n
= 9) and EPSP bursts (n
= 11) neurones. D, sample trace showing how various parameters of the first evoked spike were measured (as described in Methods). E, for each neurone, the mean value of each parameter was measured before (t
= 0–5 min) and after (t
= 20–25 min) the synaptic stimulation, and the percentage change was calculated.
