First published online December 15, 2004
Journal of Experimental Biology 208, 141-155 (2005)
Published by The Company of Biologists 2005
doi: 10.1242/jeb.01358
Dendritic spike back propagation in the electrosensory lobe of Gnathonemus petersii
Leonel Gómez1,2,*,
Morten Kanneworff2,3,
Ruben Budelli4 and
Kirsty Grant2
1 Laboratory of Neuroscience, University of the Republic, Montevideo,
Uruguay
2 Unité de Neurosciences Intégratives et Computationnelles,
CNRS, 91198 Gif-sur-Yvette, France
3 Center for Sound Communication, Institute of Biology, University of
Southern Denmark, DK-5230 Odense M, Denmark
4 Department of Biomathematics, University of the Republic, Montevideo,
Uruguay
View larger version (50K):
[in a new window]
|
Fig. 1. Experimental setup and methods. (A) A semi-schematic representation of the
ELL slice. Stim1 represents the molecular layer stimulus and
Stim2 the trans-ELL field stimulation. The dotted line represents
the alignment of successive recording sites. EGp, Eminentia Granularis
posterior; Mz, dorsal zone, Dlz, dorsolateral zone and Vlz, ventrolateral zone
of ELL. (B) Series of field potential recordings in response to molecular
layer stimulation (Stim1), made at successive sites through the
layers of ELL as indicated by the dotted line in A, starting from the deep
granular cell layer (distance 0 mm) to the outer molecular layer (distance 750
mm). See also Fig. 2 for an
explanation of cell layers. (C) Enlargement of selected traces
(Vb, V0, Va) from
B, to illustrate how current source densities (CSD) are calculated. See
Equation 3 in text.
|
|
View larger version (28K):
[in a new window]
|
Fig. 3. Method for estimating `length constant' of apical dendrites. (A) CSD in a
gray-scale map for data obtained with molecular layer stimulation. A typical
Source/Sink/Source spatial pattern can be observed (distal towards top). (B)
Spatial distribution of current density profile at time indicated by dotted
line in A, corresponding to the peak of the synaptic sink (distal towards
top). Inset in B shows the exponential function fitted to the distal tail of
the synaptic source.
|
|
View larger version (53K):
[in a new window]
|
Fig. 4. Responses to paired pulse molecular layer stimulation. (A) Left: MG cell
reconstruction to serve as a spatial reference. Right: CSD pseudocolor map
with field potential traces superimposed. Arrowheads above traces indicate the
times of stimulus application (Stim). (B) Calculation of the difference
between the first and second responses. Color bar on the right indicates
observed increase or decrease in entering current density when the first
response was subtracted from the second. Dotted line indicates the spatial
displacement and time course of the backpropagated event. Note that both the
initial sink (associated with field potential N2) and the later
sink (associated with field potential N3) are increased in response
to the second stimulus pulse.
|
|
View larger version (36K):
[in a new window]
|
Fig. 5. Responses to trans-ELL field stimulation. (A) Left: cell profile as a
spatial reference. Right: CSD as a pseudocolor map with field potential
recordings superimposed. (B) Selected CSD traces from A aligned with the
corresponding height in the pseudocolor map in A. These show that the stimulus
current peak (black dotted line) inverts (see arrows) somewhere between the
ganglionic and granular layers; thus, current enters the cells all along the
apical dendritic tree and leaves the cells close to the ventral pole, at the
level of the axon initial segment and the basal dendrites.
|
|
View larger version (61K):
[in a new window]
|
Fig. 6. Responses to multiple pulses, compared using the two stimulation methods
(molecular layer and trans ELL field) described in the text. (A) (top) Three
successive stimuli to the molecular layer produced increasing facilitation;
(bottom) three successive trans-ELL field stimuli resulted in depression of
2nd and subsequent responses. Data were acquired concomitantly by applying the
two different stimuli in alternate sweeps. (B) The third response in A (top)
is shown with an expanded timebase. Arrow shows a sink that appeared midway
between the site of synaptic entry and the ganglionic layer response, which
was observed in some, but not all, cases. Arrowheads show that the ganglionic
sink was in fact composed of two distinct sinks occurring at the same depth,
one a little earlier than the other. Both sinks had a tendency to propagate
back into the molecular layer. It was always the case that when these two
sinks appeared together, the earlier one was thinner, more elongated, and
seemed to propagate outward faster. Abbreviations, see legend to
Fig. 1.
|
|
View larger version (32K):
[in a new window]
|
Fig. 7. Intracellular recordings from an MG cell. Molecular layer stimulation
(arrows) was adjusted initially below threshold for evoking a postsynaptic
action potential and was then applied repetitively at a frequency
representative of the natural electric organ discharge (EOD) rhythm during
increased sensory attention. Postsynaptic responses increased with the number
of stimuli, leading to the generation of partial (arrowhead) and then full
backpropagating broad spikes (asterisk), and to increasing numbers of small
narrow spikes (probably axon spikes). Note that a slow depolarization, lasting
several hundred ms, builds up with repetitive stimulation. The inset (top
right) shows a plot of the probability of evoking a broad spike as a function
of the number of stimuli applied.
|
|
View larger version (63K):
[in a new window]
|
Fig. 8. Blocking of backpropagation by TTX. (Left) Diagram of experimental setup
showing position of recording electrodes in outer molecular layer (M) and
ganglionic layer (G). Green spot and shaded area represent site of TTX
application in the deep fiber layer and its diffusion outwards. (Right) Color
representation of field potentials obtained from M (top) and G (bottom)
recording points. N2 and N3 are indicated in the upper
panel. Dotted line indicates the moment at which TTX was applied. The
ganglionic layer response disappeared 45 s after TTX application and
simultaneously N3 disappeared in the molecular layer response.
N2 persisted for a further 1.5 min.
|
|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati 
Twitter What's this?
© The Company of Biologists Ltd 2005
1 and