| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Cellular and Molecular Medicine, University of Ottawa, Ottawa, Ontario K1H 8M5, Canada
 |
ABSTRACT |
|---|
 Abstract
 Introduction
Methods
Results
Discussion
References
|
|---|
Wang, Daliang and Leonard Maler. In vitro plasticity of the direct feedback pathway in the electrosensory system of Apteronotus leptorhynchus. J. Neurophysiol. 78: 1882-1889, 1997. We have used field and intracellular recording from pyramidal cells in an in vitro preparation of the electrosensory lateral line lobe (ELL) of Apteronotus leptorhynchus to investigate synaptic plasticity of a direct feedback pathway: the (StF). Tetanic stimulation of the StF enhanced the StF-evoked synaptic response by 145% in field and the excitatory postsynaptic potential (EPSP) 190% in intracellular recordings. Maximal enhancement occurred at 5 s and lasted for ~120 s. Tetanic frequencies of 100-300 Hz produced enhancement; lower or higher frequencies failed to produce statistically significant changes in EPSP amplitude. Rates of 100-200 Hz occur in vivo in the cells of origin of the StF, suggesting that this form of plasticity may be operative under natural conditions. We could not elicit either long-term potentiation or depression by any stimulation protocol of the StF; in the case of long-term potentiation, this held even when excitatory transmission was enhanced by application of bicuculline, a 
-aminobutyric acid-A antagonist. When tetanic stimulation of the StF was paired with hyperpolarization of pyramidal cells, subsequent StF-evoked EPSPs were increased by 146% (5 min posttetanus); this anti-Hebbian synaptic enhancement lasted for ~10 min. Neither tetanic stimulation alone, hyperpolarization alone, nor tetanic stimulation paired with pyramidal cell depolarization altered StF-evoked EPSP amplitudes on this time scale. Anti-Hebbian synaptic enhancement was not blocked by the N-methyl-D-aspartate-receptor antagonist D.L-aminophosphovalerate. The in vitro demonstration of anti-Hebbian plasticity at StF synapses replicates similar in vivo results. Anti-Hebbian synaptic plasticity of the StF may be responsible in part for the ability of gymnotiform fish to reject redundant electrosensory signals.
Sensory systems signal the occurrence of events significant to the organism. To do so, they must extract relevant features under varying conditions. Obvious examples include the ability of the visual system to identify objects under a very wide range of background illumination and the auditory system's remarkable capacity to detect low-level sounds in the presence of noise. This is partly due to the capacity of these systems to reject unchanging or irrelevant input. This ability is also prominent in the electrosensory system. For example, the electroreceptors in elasmobranches are stimulated strongly by the repetitive low-frequency electric signals produced by their own respiration. In the CNS, this input is filtered out, permitting the animal to detect relatively weak exogenous signals (Montgomery and Bodznick 1994 Slice preparation
The fish were anesthetized with MS-222 (Sigma, 1:15000), respirated (aerated water), and brain slices through the ELL prepared as previously described (Berman et al. 1997 Stimulation
Stimulation was delivered via a unipolar tungsten electrode placed on the dorsal part of the StF in the medial segment of the ELL as previously described (Berman et al. 1997 Extracellular recordings
Field potentials were recorded from the VML of the centromedial segment (CMS) of the ELL (Berman et al. 1997 Intracellular recordings
Intracellular recordings were made from CMS pyramidal cells with glass micropipettes pulled on a Brown-Flaming puller (Model P-87, Sutter Instrument, CA), filled with 3 M potassium acetate and bevelled to a resistance of 70-130 M Application of pharmacological agents
Drugs (from Sigma unless otherwise stated) were administered by bath application: 4 mM MnCl2 in ACSF (with 0.02 mM CaCl2), 70 nM bicuculline methochloride (dissolved in distilled water at 0.07 mM and further diluted in ACSF before use) and 100 nM D.L-aminophosphovalerate (APV, dissolved in NaOH at 0.1 mM and further diluted in ACSF before use).
Data analysis
Data analysis was done with IgorPro (Wavemetrics) and Statistica. Statistical significance was assessed by either Student's t-tests or analysis of variance (ANOVA).
The ELL has four segments: medial (MS), CMS, central lateral (CLS) and lateral (LS). Electroreceptors of gymnotiform fish are of the ampullary or tuberous type: ampullary receptors respond to low-frequency exogenous electric fields, whereas tuberous receptors are tuned to the fish's own EOD frequency. The MS receives input from ampullary receptors, whereas the CMS, CLS, and LS receive tuberous input (Carr and Maler 1986 Transient synaptic enhancement: posttetanic potentiation
FIELD POTENTIAL RECORDINGS.
As previously described (Berman et al. 1997
INTRACELLULAR RECORDINGS.
Electrophysiological characteristics of pyramidal cells (n = 26) were: input resistance, 34.1 ± 1.0 M Lack of LTP of StF synapses in VML
To maximize the possibility of LTP of the StF pathway, we used stronger stimuli and five tetani (see METHODS) in these experiments. Tetanization at 100 Hz as well as at other frequencies (data not shown) did not produce statistically significant LTP (n = 10) in field recordings (Fig. 2). The percentage increase during 60 min was 106%, which did not differ significantly from control values. ANOVA demonstrated that, with this more intense stimulation, there was a significant potentiation (118%) at 10-min posttetanus (P < 0.05) but not at longer delays. In other preparations, the inability of tetanic stimulation to induce LTP has been attributed to excessive
Anti-Hebbian plasticity in the StF pathway in vitro
Bastian (1996b) The present study demonstrates that tetanic stimulation of the StF pathway produces two types of synaptic plasticity of ELL pyramidal cell EPSPs: first, a short-lived (~2 min) PTP; second, a longer-lasting (~10 min) anti-Hebbian potentiation when the tetanic stimulation is paired with postsynaptic hyperpolarization. However, our stimulation paradigms did not induce any long-term change (>30 min; LTP or LTD) in synaptic efficacy.
INTRODUCTION
Abstract
Introduction
Methods
Results
Discussion
References
; see also Nelson and Paulin 1995). Bastian (1995, 1996a,b) recently has demonstrated a similar filtering mechanism in the gymnotiform fish Apteronotus leptorhynchus. These fish often bend their tails in a rhythmic fashion and the amplitude of their electric organ discharge (EOD) alternately increases and decreases globally over each side of their body in phase with this movement. The magnitude of the EOD changes is sufficient to drive electroreceptors, which respond in phase with increases in EOD amplitude. Recordings by Bastian in the electrosensory lateral line lobe (ELL), revealed that pyramidal cells failed to respond to this repetitive AM. Further experiments demonstrated that this was due to an active cancellation mechanism that operated in an anti-Hebbian manner: the response to any local input paired in phase with tail bending was diminished due to the development of a "negative image" of the input as defined by Bell (1981, 1982); response to local input in antiphase with tail bending increases due to a negative image of opposite sign to the input.
,b) was able to demonstrate that the negative image was produced, at least in part, by glutamatergic feedback fibers to the ELL molecular layer, and he ascribed some portion of this effect to the stratum fibrosum (StF) projection to the ventral molecular layer (VML). Our experiments in an in vitro slice preparation of the ELL have characterized the properties of StF-evoked excitatory postsynaptic potentials (EPSPs) and demonstrated that they have a voltage-dependent component partly mediated by N-methyl-D-aspartate (NMDA) receptors (Berman et al. 1997). We report in this paper that, under the controlled conditions of a brain slice, we can replicate half of the anti-Hebbian plasticity that Bastian has described in vivo: pairing hyperpolarization of ELL pyramidal cells with tetanic stimulation of the direct feedback pathway enhances the response to subsequent stimulation of this pathway. The direct feedback pathway appears, however, to lack the ability to support the more classic forms of synaptic plasticity, long-term potentiation and depression (LTP, LTD), although short-term synaptic enhancement is elicited readily. The ability to produce anti-Hebbian plasticity in an in vitro preparation may facilitate a molecular dissection of this type of synaptic plasticity.
METHODS
Abstract
Introduction
Methods
Results
Discussion
References
; Turner et al. 1994). The slices (300-400 µm thick) were placed in an interface chamber at room temperature (~23°C) so that the rostral face of the ELL was uppermost and perfused with oxygenated (95% O2-5% CO2) artificial cerebrospinal fluid [ACSF, containing (in mM) 124 NaCL, 24 NaHCO3, 10 D-glucose, 1.25 KH2PO4, 2 KCL, 2 MgSO4, and 2 CaCl2] at a flow rate of 2.5-3 ml/min. Recording were initiated after a 60- to 90-min recovery period.
). Stimulus pulses were 20-100 µs in duration with an intensity that produced two-thirds of the maximal EPSP amplitude (30-50 V for field and3-10 V for intracellular recording); in the cases where we attempted to elicit LTP, we used larger stimulus intensities that produced maximal EPSPs. We attempted to produce alterations in StF-evoked EPSPs by using stimulus trains of 10 pulses at frequencies of 1-350 Hz delivered three times at intervals of 1-3 s; for the LTP cases, we delivered the tetani five times.
). Baseline field EPSPs were collected for 
15 min before tetanic stimulation; each pretetanus trial was the average of 20 recordings. We only initiated further experiments in cases where the baseline recordings produced stable EPSPs (mean change <5%). In experiments designed to examine transient effects of tetanus, posttetanic recordings were initiated 5 s after the tetanus to avoid contamination by long-lasting potentials [either plateau depolarizations or inhibitory postsynaptic potentials (IPSPs) often >1 s] produced by tetanic stimulation. StF-evoked EPSPs were recorded at 5-s intervals for 2 min; because short-term enhancement decayed rapidly, no averaging was done for the posttetanic recordings. In experiments designed to study possible long-term plasticity (LTP or LTD), recording trials were collected every 5 min for 60-90 min; 20 EPSPs were averaged at each test interval.
(Model BV-10, Sutter Instrument); the electrophysiological properties of the impaled cells (input resistance, spike width) (Berman et al. 1997; Turner et al. 1994) suggested that we record from pyramidal cells, which are the main cell type in this layer (Maler 1979). Input resistance was monitored by injection of hyperpolarizing current pulses. Further technical details of the recording setup can be found in Berman et al. (1997).
RESULTS
Abstract
Introduction
Methods
Results
Discussion
References
). The overall morphology of the three tuberous segments is similar, although subtle differences related to spatial and temporal processing of electrosensory signals have been described (summarized in Turner et al. 1996). The ELL is a laminar structure (see Berman et al. 1997): primary afferents (deep fiber layer) terminate in a neuropil layer on the basal dendrites of pyramidal cells (projection neurons) and various interneurons. Pyramidal cells and interneurons are located in separate pyramidal cell and granular layers. Apical dendrites of the pyramidal cells and some interneurons ramify in a large molecular layer where they receive feedback input. The VML receives input from the n. praeminentialis via the StF (Maler et al. 1982), whereas the dorsal molecular layer (DML) contains parallel fibers from overlying cerebellar granule cells (Maler 1979). The StF fibers terminate on pyramidal cell dendritic spines (Maler et al. 1981); this projection previously has been shown to be glutamatergic (Wang and Maler 1994) and uses ionotropic (NMDA and non-NMDA) receptors (Berman et al. 1997).
), stimulation of the StF produced a characteristic biphasic field potential in the VML (Fig. 1): a small initial positivity followed by a larger negativity (in some cases the positivity is obscured by the stimulus artifact). The negativity was typically 0.5-2 mV in amplitude, depending on stimulus intensity, and peaked with a latency of ~6 ms (6.2 ± 0.4 ms; mean ± SE). This negativity reflects primarily StF-evoked EPSPs in pyramidal cell proximal apical dendrites (Berman et al. 1997). In control experiments (single pulse test stimulation without delivery of tetanic stimulation), the field EPSP remained stable for 60 min (Figs. 1 and 2B).
View larger version (37K):
[in a new window]
FIG. 1.
A1, top: superimposed stratum fibrosum (StF)-evoked field potentials before (1) and 5 s after (2) tetanic stimulation; bottom: subtracted response (2 
1) reveals potentiated field potential. A2: averaged response of tetanically stimulated (TS; n = 11) and control (con; n = 13) StF-evoked field excitatory postsynaptic potentials (EPSPs; bars represent SE). Note that the pretetanus time scale is in minutes (15 min of baseline EPSPs were collected), and the posttetanus time scale is in seconds. Inset: averaged response of tetanically stimulated StF-evoked EPSPs (error bars omitted for clarity; 
, 5-s point) over a longer baseline; note decline of potentiated EPSP to baseline during 10 min. B1, top: superimposed StF-evoked field potentials before (1) and 5 s after (2) tetanic stimulation in Mn2+ artificial cerebrospinal fluid (ACSF); bottom: subtracted response (2 1) reveals lack of potentiation of StF fiber volley. Inset: reduction of StF-evoked field potential in Mn2+ ACSF. B2: averaged response of tetanically stimulated StF fiber volley in Mn2+ ACSF (n = 11). C1: tetanic stimulation of StF produces a characteristic augmenting compound EPSP in intracellular recording. C2, top: superimposed StF-evoked intracellular EPSP before (1) and 5 s after (2) tetanic stimulation; bottom: subtracted response (2 1) reveals potentiated EPSP. C3: averagedr e s p o n s e o f t e t a n i c a l l y s t i m u l a t e d(n = 10) StF-evoked EPSP. Inset: averaged response during 3 min recording time (error bars omitted for clarity; , 5-s point; note decline of potentiated EPSP to baseline in <3 min).
View larger version (18K):
[in a new window]
FIG. 2.
A: Bicuculline (bic) enhances the StF-evoked field EPSP in both the ventral molecular layer (VML; A1) and pyramidal cell layers (PCL; A2) over control values (con). Tetanic stimulation of StF in control ACSF or in the presence of bicuculline does not cause long-term potentiation of the StF-evoked field EPSP in either the VML (B1) or PCL (B2).
): stimulation was 100 Hz for 100 ms (10 pulses) given three times at 1-3 s intervals. This stimulation protocol produced an increase in the field EPSP that was maximal 5 s posttetanus (peak: 144.8 ± 9.9%, n = 11; Fig. 1); this was significantly greater than the time-matched control (104.9 ± 1.1%, n = 13, P < 0.01). The average enhancement during 2 min posttetanus was 120.4 ± 24.1%, which was also significantly different from the control (100.8 ± 2.9%, P < 0.02, Fig. 1). The potentiation decayed during a period of 1-10 min (Fig. 1A2, inset); two-way ANOVA demonstrated that the EPSP was not significantly different from pretetanus levels after 105 s (least significant digit post hoc test at 105 s, P < 0.05; a more stringent Tukey test gives significance only for the 5- and 10-s points). The time course of this enhancement is compatible with the expression of either augmentation and/or posttetanic potentiation (PTP), two commonly observed forms of transient enhancement of synaptic transmission after tetanic stimulation (Magleby 1979; Zucker 1989, 1996). For the sake of simplicity, we refer to the enhancement as PTP because we do not have kinetic evidence that it is due to more than one molecular mechanism (Zucker 1996).
) and still permits visualization of individual EPSPs during intracellular recordings. Stimulation frequencies of 1, 10, 50, and 350 Hz (10 pulses) did not produce statistically significant PTP; notably the 1-Hz stimulation did not produce LTD at these synapses (data not shown).
). In the presence of Mn2+, the field amplitude was reduced to ~60% of control values (n = 11; Fig. 1). The same tetanic stimulation protocol (100 Hz) did not produce any change in the amplitude of the fiber volley (Fig. 1). The PTP we observe is therefore due to potentiation of the StF-evoked synaptic potential.
; RMP, 61.3 ± 0.8 mV; and current-evoked action potentials, 73.0 ± 4.0 mV.
). Our stimulation site was chosen to reduce activation of direct inhibitory feedback fibers (Maler and Mugnaini 1994), and only small or no IPSPs usually were evoked. Tetanic stimulation produced a maximum enhancement of the EPSP at 5-s posttetanus (191.4 ± 15.3%, n = 10); averaged during 2 min the EPSP was increased by 141% (140.6 ± 9.4%, Fig. 1). This was substantially larger than the potentiation observed with field recordings; the difference is likely due to the fact that the field EPSP was contaminated by the StF fiber volley (Berman et al. 1997), which was not affected by tetanic stimulation (see above). The larger value (190% peak increase) is therefore likely to better reflect the true extent of PTP in this system. The latency to the peak of the EPSP did not change at the time of maximal potentiation nor was there any statistically significant alteration in the late phase of the EPSP (at 20-ms latency, Fig. 1). The amplitude of the EPSP decayed back to pretetanus values by 2 min (Fig. 1; least significant digit post hoc test, P < 0.005 at 120 s; the more stringent Tukey test reveals significance for only the 5- and 10-s points). During the expression of PTP, there was no change of pyramidal cell input resistance and the resting membrane potential returned to baseline by ~1 s. The intracellular results therefore confirmed our conclusion from field recordings: StF-evoked EPSPs show prominent PTP when activated by physiologically appropriate tetani, and this synaptic enhancement is not associated with alteration of the electrophysiological properties of pyramidal cells.
-aminobutyric acid-A (GABAA) inhibition (Artola and Singer 1987; Stevens and Cotman 1991; Steward et al. 1990). Bath application of bicuculline (n = 11) increased the field EPSP by 30% in both VML and the pyramidal cell layer (Fig. 2), as expected given the presence of GABAergic interneurons, which receive input from StF (Maler and Mugnaini 1994). In the presence of bicuculline, tetanic stimulation increased the mean EPSP amplitude by 115% in VML (115.6 ± 5.9%, n = 11, Fig. 2) and by 113% in the pyramidal cell layer (113.1 ± 9.9%, n = 11, Fig. 2) during the 60-min posttetanic period. This was not significantly different from either the mean pretetanic EPSP or that produced by 100-Hz tetanic stimulation without bicuculline (106.3 ± 3.6%, n = 10). In the presence of bicuculline, there was significant potentiation at 10 min (120%, P < 0.005) and 20 min (P < 0.02) posttetanus, but not after longer delays.
View larger version (33K):
[in a new window]
FIG. 3.
A1: pairing tetanic stimulation of StF with hyperpolarization of a pyramidal cell (n = 16) causes an augmenting compound EPSP that fails to reach spike threshold. A2, top: superimposed StF-evoked EPSP before (1) and 5 min after (2) tetanic stimulation paired with hyperpolarization; bottom: subtracted response (2 1) reveals potentiated EPSP. B1: pairing tetanic stimulation of StF with depolarization of a pyramidal cell (n = 15) evokes action potentials (truncated). B2, top: superimposed StF-evoked EPSP before (1) and 5 min after (2) tetanic stimulation paired with depolarization; bottom: subtracted response (2 1) reveals minimal alteration of the EPSP. C: averaged long-term effect of the pairing paradigms (hyp + ts, hyperpolarization + tetanus; dep + ts, depolarization + tetanus) and tetanic stimulation alone (ts) on StF-evoked EPSPs. Tetanic stimulation paired with hyperpolarization produces significant enhancement of the StF-evoked EPSP at 5 and 10 min posttetanus. Other protocols do not produce a significant alteration in the StF-evoked EPSP.
recently has demonstrated that in vivo pairing of tetanic stimulation of StF with hyperpolarization of pyramidal cells produced an enhancement of StF evoked EPSPs; this change in synaptic efficacy is of far longer duration (>10 min) than that of PTP (<2 min with moderate stimulation). Pairing StF tetanus with depolarization produces little alteration of the EPSPs but increases a StF-evoked hyperpolarization. We therefore replicated Bastian's protocol in vitro to determine whether this form of plasticity was maintained in the isolated ELL.
) that always remained below spike threshold (Fig. 3). Tetanic stimulation during depolarization produced EPSPs that triggered action potentials (Fig. 3). We were concerned that possible long-term alterations in StF transmission might decrement on testing (Bastian 1996a). We therefore collected posttetanic data at a single time after tetanus (at 5- to 20-min intervals posttetanus). In some experiments, we simultaneously recorded field potentials in the StF to monitor the stability of the fiber volley.
has suggested that pairing depolarization and tetanus might enhance StF-evoked IPSPs; we therefore measured the amplitude of the StF-evoked synaptic response at a latency of 20 ms (the IPSP is near maximal at this time) (N. J. Berman, personal communication); there was no significant alteration from that of control or other conditions (0.7 ± 0.6 mV, n = 15, P > 0.5).
; Coan et al. 1987; Collingridge et al. 1983). APV application reduced StF-evoked EPSPs (data not shown) as previously shown (Berman et al. 1997). Tetanization of StF paired with hyperpolarization of pyramidal cells still produced an increase of the StF-evoked EPSP of 146% (146.5 ± 10.3%, n = 4) at 5 min posttetanus. This potentiation was not significantly different from that observed in the drug-free tetanization-hyperpolarization group (see above). We conclude that anti-Hebbian potentiation of the StF does not depend on NMDA receptors.
DISCUSSION
Abstract
Introduction
Methods
Results
Discussion
References
); PTP has been analyzed quantitatively at invertebrate synapses and the vertebrate neuromuscular junction where quantal analysis indicates that it is presynaptic (Atwood and Wojtowicz 1986; Magleby 1979; Zucker 1989, 1996). Similar degrees of PTP after StF stimulation also has been observed in vivo (~120%) (Bastian 1996b). Pharmacological experiments suggest that, as in other preparations, PTP of the StF is probably a presynaptic phenomenon (unpublished observations).
), a hypothesis supported by the anatomy (Maler and Mugnaini 1994) and physiology (Berman et al. 1997) of this system. The presence of PTP at StF synapses in VML further suggests that the putative searchlight can be modified by the recent experience of the fish. Because A. leptorhynchus scans objects (e.g., worms) in a stereotyped manner (Lannoo and Lannoo 1993), it is likely that an object will stimulate the same sequence of receptors during successive scans, leading to PTP of StF synapses and an improved ability to locate that object for the next minute; the improvement will be most marked (~200%) for 5-10 s. Clearly, given the complex dynamics of this feedback pathway (Berman et al. 1997), these ideas will have to be modeled and tested by in vivo experiments.
; Bliss and Lømo 1973); stimulation of these same fibers at low rates produces LTD (Dudek and Bear 1992; Dunwiddie and Lynch 1978; see Bear and Abraham 1996 for review). Both the LTP and LTD at these sites depend on NMDA receptors and Ca2+ influx (Malenka 1991; Mulkey and Malenka 1992). Similar NMDA-receptor-mediated long-term forms of synaptic plasticity are found in cortex (Artola and Singer 1987; Hirsch and Crepel 1990; Tsumoto et al. 1987). StF fibers use glutamate as a transmitter (Wang and Maler 1994) and StF-evoked EPSPs are mediated in part by NMDA receptors (Berman et al. 1997). It was therefore surprising that we could not elicit either LTP or LTD with any stimulation protocol or during pharmacological blockade of inhibition. LTP and LTD depend on a complex cascade of second messenger interactions including postsynapticcalcium/calmodulin dependent kinase II, 
isoform (CaMKII) (Glazewski et al. 1996; Ito et al. 1991; Malenka et al. 1989; Malinow et al. 1989; Silva et al. 1992). Although CaMK2 is present in ELL, the isoform is confined to StF terminals in VML, whereas pyramidal cells contain the CaMK2
isoform (L. Maler and M. Hincke, unpublished observations). We therefore hypothesize that it is not possible to elicit LTP from ELL pyramidal cells (via any excitatory input) because they lack a critical second messenger system: CaMKII. This appears reasonable from the standpoint of sensory processing, because LTP or LTD of StF-evoked EPSPs would prevent the electrosensory system from tracking rapidly changing input.
paired electrical stimulation of the StF with either physiological activation of ELL pyramidal cells via input to their receptive fields or current injection into these cells to hyper- or depolarize them. The latter experiments are directly comparable with ours, and their results are in close agreement. In vivo pairing of tetanic stimulation of StF with hyperpolarization of pyramidal cells led to a significant increase in the amplitude of StF-evoked EPSPs of 143.3% (Bastian 1996b) (our result was 145.6%). Our replication of Bastian's results in vitro confirms that anti-Hebbian potentiation can be triggered by nonpathological stimulation of StF fibers.
reported that pairing tetanic stimulation of StF with depolarization of pyramidal cells caused subsequent StF stimulation to evoke a hyperpolarization of pyramidal cells and reduce their spike discharge; this is the other side of anti-Hebbian plasticity. Equivalent pairings in our slice preparation did not produce such hyperpolarizations: anti-Hebbian plasticity in the slice is one-sided. Bastian has proposed that the hyperpolarization in his experiments was due to an enhancement of StF-evoked inhibition. It is therefore possible that, in our in vitro preparation, we were not effectively activating feedback inhibitory fibers and/or inhibitory interneurons (Maler and Mugnaini 1994).
; Reyes and Stanton 1996; Sakurai 1990; see Ghosh and Greenberg 1995 for review). Our present results appear paradoxical and suggest that novel mechanisms may underlie anti-Hebbian plasticity in the ELL StF pathway.
,b). Its role is presumably to reduce the responsiveness of the electrosensory system to expected reafferent input such as those caused by rhythmic tail bending (Bastian 1996b). Anti-Hebbian plasticity is a common feature of electrosensory systems (Bell 1981; Montgomery and Bodznick 1994) and the mammalian cerebellum (Crepel and Jaillard 1991). In these cases, it is also likely to be caused by anti-Hebbian alterations of input to the molecular layer (Bell et al. 1993; Nelson and Paulin 1995).
). ELL pyramidal cells may hyperpolarize if there is a tonic decrease in their input [basilar pyramidal cells or E cells; the opposite would be expected for nonbasilar or I cells (Saunders and Bastian 1984)], and this would reduce their response to a local increase in EOD amplitude. If anti-Hebbian plasticity increased the response of the StF input to the pyramidal cell, it would compensate for the hyperpolarization; hence pyramidal cell response to local input would become independent of the cell's mean membrane potential. The StF then might be considered to operate at three time scales. Anti-Hebbian plasticity normalizes StF-evoked EPSPs during minutes, allowing the system to cancel out slow changes in EOD amplitude. PTP, superimposed on the anti-Hebbian plasticity, enhances the response to StF input during a few seconds. The voltage-dependent augmenting response of pyramidal cells to StF input (Berman et al. 1997) is superimposed on PTP and increases the effectiveness of the putative "searchlight" mechanism over a millisecond time scale.
Several recent studies have emphasized the importance of normalizing synaptic strength (Abbott et al. 1997; Bell and Sejnowski 1995; De Schutter 1995). Although the underlying cellular mechanisms are probably different, in each case, normalization permits the output cell to respond in an appropriate fashion over a wide range of afferent input. In particular, Abbott et al. (1997) have emphasized the role of synaptic depression in normalizing gain control of afferent inputs discharging tonically at different rates. Synaptic depression would be inappropriate for the StF, a system that discharges phasically (Bratton and Bastian 1990) and may require an augmenting synaptic response (Berman et al. 1997). A comparative analysis of synaptic normalization may better reveal how biophysical mechanisms are adapted to the requirements of different neural systems.
| |
ACKNOWLEDGEMENTS |
|---|
We thank Drs. J. Bastian and N. Berman for helpful discussions and W. Ellis for technical support.
This work was supported by a Medical Research Council Grant toL. Maler.
| |
FOOTNOTES |
|---|
Address reprint requests to L. Maler.
Received 7 April 1997; accepted in final form 18 June 1997.
| |
REFERENCES |
|---|
|
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
This article has been cited by other articles:
|
|
 |
|
  A.-M. M. Oswald, J. E. Lewis, and L. Maler Dynamically Interacting Processes Underlie Synaptic Plasticity in a Feedback Pathway J Neurophysiol, May 1, 2002; 87(5): 2450 - 2463. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Bastian and J. Nguyenkim Dendritic Modulation of Burst-Like Firing in Sensory Neurons J Neurophysiol, January 1, 2001; 85(1): 10 - 22. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Berman and L Maler Neural architecture of the electrosensory lateral line lobe: adaptations for coincidence detection, a sensory searchlight and frequency-dependent adaptive filtering J. Exp. Biol., January 5, 1999; 202(10): 1243 - 1253. [Abstract] [PDF]
|
|
|
|
|
|
J Bastian Plasticity of feedback inputs in the apteronotid electrosensory system J. Exp. Biol., January 5, 1999; 202(10): 1327 - 1337. [Abstract] [PDF]
|
|
|
|
|
|
C. Bell, V. Han, Y Sugawara, and K Grant Synaptic plasticity in the mormyrid electrosensory lobe J. Exp. Biol., January 5, 1999; 202(10): 1339 - 1347. [Abstract] [PDF]
|
|
|
|
 |
|
 D. Wang and L. Maler Differential roles of Ca2+/calmodulin-dependent kinases in posttetanic potentiation at input selective glutamatergic pathways PNAS, June 9, 1998; 95(12): 7133 - 7138. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Bastian Plasticity in an Electrosensory System. III. Contrasting Properties of Spatially Segregated Dendritic Inputs J Neurophysiol, April 1, 1998; 79(4): 1839 - 1857. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
