First published online March 2, 2006
Journal of Experimental Biology 209, 1122-1134 (2006)
Published by The Company of Biologists 2006
doi: 10.1242/jeb.02080
The role of single spiking spherical neurons in a fast sensory pathway
Javier Nogueira1,2,
María E. Castelló1,2 and
Angel Ariel Caputi1,*
1 Department of Integrative and Computational Neurosciences, Instituto de
Investigaciones Biológicas Clemente Estable, Associated Unit of the
Facultad de Ciencias, Universidad de la República, Av. Italia 3318,
Montevideo, Uruguay
2 Department of Histology and Embriology, Facultad de Medicina, Universidad
de la República, Gral. Flores 2515, Montevideo, Uruguay
View larger version (23K):
[in a new window]
|
Fig. 1. The low-responsiveness window of the fast electrosensory pathway is
elicited by natural (A) and artificial (B–D) stimuli. (A) Field
potentials evoked by the self-(black triphasic artifact around time zero in
each trace) and conspecific (green)-generated EODs at the magnocellularis
nucleus in a freely moving fish. The triphasic waveform at the beginning of
each trace (labeled 1–5) is the sEOD artifact, followed by a spike
corresponding to the fast electrosensory pathway evoked response. The cEOD
artifact is the small triphasic waveform highlighted in green (labeled
a–g). The fast electrosensory pathway response evoked by the
conspecific-generated EOD (green spikes) is absent at short delays (a and f)
and increases in amplitude with the interval between the self-generated EOD
and the conspecific-generated EOD (green evoked spikes labeled e, d, c and b).
Note that a similar, but smaller, decrease in the response to the
self-generated EOD is provoked by the activation of the fast electrosensory
pathway when the conspecific-generated EOD (second trace, blue spike) occurs
just before the self-generated EOD (red spike). (B) Field potential responses
evoked in the magnocellularis nucleus of a curarized fish by a two-threshold
artificial stimulus delivered at different periods (28–100 ms) after a
conditioning stimulus (evoking a response similar to that evoked by the
self-generated EOD). With short delays after the activation of the fast
electrosensory pathway responsiveness is reduced, as indicated by a decrease
in amplitude (C) and an increase in latency (D) of the response elicited by
the artificial test stimulus.
|
|
View larger version (45K):
[in a new window]
|
Fig. 2. The increase of the magnocellularis response with the intensity of the test
stimuli or the conspecific-generated EOD shows that the low-responsiveness
window is not an all-or-none phenomenon. (A) Amplitude of the magnocellularis
nucleus responses to artificial test stimuli of different intensities as a
function of the delay between the self-generated EOD and the test stimuli. The
responses are expressed as a percentage of the maximal response evoked by the
self-generated EOD. (B,C) Superimposed traces triggered by the self-generated
EOD obtained in a fish chronically implanted in the magnocellularis nucleus,
showing several responses evoked by the conspecific-generated EOD of a fish of
similar length (13 cm) placed antiparallel (B) 6 cm and (C) 3 cm away.
|
|
View larger version (25K):
[in a new window]
|
Fig. 3. The responsiveness of the fast electrosensory pathway decreases with the
intensity of the conditioning stimuli. To assess the effect of activation of
the fast electrosensory pathway on its responsiveness, the amplitude of the
response of the magnocellularis nucleus to an artificial test stimulus of
intensity 2x the threshold was studied as a function of the amplitude of
an artificial conditioning stimulus, and the delay between conditioning and
test stimuli, in a curarized fish. The amplitude of the test response (as a
percentage of its maximal response) is plotted as a function of the delay
between conditioning and test stimuli for eight conditioning stimulus
intensities (0–10x threshold).
|
|
View larger version (43K):
[in a new window]
|
Fig. 4. The spherical cells `phenotype'. (A,B) Biocytin-filled spherical cell
characteristically firing a single spike when stimulated by long-lasting
current step. (A) The spherical cells bear round smooth soma of about 15 µm
diameter from which only one thin dendrite might emerge (not in this case) and
a thick axon with a thin initial segment. (B) Traces from the same cell are
displayed at a different magnification to show the shapes of the spikes (red
and black traces) and the absence of repetitive firing with long lasting
stimuli (inset). (C) Some cells (black and red traces) fire the spike on the
falling face of a hump evoked by the stimulus step (stimuli in nA: violet
0.95; blue 0.75, green 0.60, black 0.50, red 0.40). (D) Spike latency was a
hyperbolic function of the stimulus intensity (data from another cell).
|
|
View larger version (21K):
[in a new window]
|
Fig. 5. Typical early subthreshold responses in a spherical cell. (A) Depolarizing
and hyperpolarizing stimuli provoke very asymmetric responses. Note the
hump-and-hold profile for depolarizing steps and a sag depolarization after 63
ms for hyperpolarizing steps (arrows in A, further analyzed in
Fig. 6). (B) Enlarged version
of the first 11 ms of the responses (boxed yellow in A, showing a hump peaking
2.4 ms after the stimulus onset (current steps are equally spaced by 25 pA).
Up to this peak (red line in B) there is a linear relationship between
membrane potential and injected current (red symbols in D). After this moment
the response is asymmetric. For depolarizing steps there is a reduction of the
voltage drop caused by the flow of injected current across the membrane and
consequently an increase in membrane conductance. As shown by the coincidence
of the depolarizing limiting slopes (broken line in D) of the
V/I plots constructed for 10.4 ms (blue line in B; blue
symbols in D) and 63 ms (green line in A; green symbols in D), there are no
further changes in the V/I slope, indicating that the
activated conductance does not inactive with time. For hyperpolarizing steps
the semilogarithmic plot (C) of the voltage derivative vs time of the
bottom voltage trace of B illustrates that there is a simple exponential
relationship (red line) up to 2.4 ms (arrows in B,C). Beyond this time there
is an upward departure from a simple exponential curve, indicating a reduction
of membrane conductance.
|
|
View larger version (24K):
[in a new window]
|
Fig. 6. Late responses to subthreshold pulses. Responses obtained from another cell
to current steps of (A) 200 ms and (B) 50 ms. Current steps are equally spaced
by 33 pA. For depolarizing currents, the long-lasting steps provoke an outward
rectification, indicated by the reduction of the spacing between voltage
recordings after the hump (A,B) and also by the comparing the
V/I relationship obtained at 54 ms (green line in A) and 195
ms (red line in A) corresponding to red and green symbols fitted by the dotted
line in D. For hyperpolarizing steps, beyond 54 ms after the onset (green line
in A) there is an inward rectification with a sag depolarization (arrows in
A), suggesting an increase of membrane conductance. Consequently, at the peak
of the hyperpolarization the limiting slope for hyperpolarizing currents is
maximal (green symbols fitted by the continuous line in D). At the end of the
current step, the return curves for depolarizations have a much faster decay
than for hyperpolarizations. When the hyperpolarizing current steps are ended
at 50 ms (B), the return curve last much longer than when steps are ended at
200 ms (A). This matches the drop in the limiting slope of the hyperpolarizing
side of the V/I plot (compare the continuous line and red
circles in D). After the long pulse, there is a rebound graded with the amount
of hyperpolarization as shown in C (enlarged version of the yellow shaded area
in A).
|
|
View larger version (47K):
[in a new window]
|
Fig. 7. Spherical cells have a long refractory period. Systematic stimulation of
the cell with pairs of intracellular steps shows that the threshold and
latency of the second spike depends on the delay between pulses. (A)
Superimposed traces obtained stimulating with four different test step
intensities (370, 380, 390 and 400 pA) at different delays from a conditioning
step (800 pA). (B) Spike latency vs amplitude of the test step
plotted for four different delays after a conditioning spike (data from
another cell). (C) Spike latency vs delay plot corresponding to the
experiment represented at the bottom row in A.
|
|
View larger version (13K):
[in a new window]
|
Fig. 8. Membrane conductance and the excitability of the cell change concomitantly
after a conditioning spike. The change in membrane resistance was evaluated by
the slope of the V/I plot constructed with data obtained 1
ms after the onset of a series of test current steps of different intensity
(dotted line in inset) applied at different delays after a conditioning spike
in four cells (symbol coded). (A) The increase in slope as a function of the
delay indicates that after the conditioning spike there is an increase in
membrane conductance and a slow return to resting values. (B,C) Change in
spike threshold and spike latency as a function of the slope. Change in
threshold was defined as the difference between the current intensity required
for eliciting a spike when the stimulus was preceded by a conditioning spike
at the given delay minus the current intensity required for eliciting a spike
without a conditioning spike. As the spike threshold and latency decrease with
the slope, the excitability of the cell is reduced by the increase of membrane
conductance caused by the conditioning spike.
|
|
View larger version (21K):
[in a new window]
|
Fig. 9. Refractoriness is graded with the conditioning step intensity. (A)
Responses evoked by four intracellular test steps (600 pA, 5 ms) applied at
four different delays after 3 ms conditioning square steps of four different
amplitudes 0, 200, 400 and 600 pA (i–iv, respectively). The increase in
conditioning step amplitude cause: (a) depression of the hump in the
subthreshold responses (compare black traces); (b) increase in spike latency
(compare red spikes with the reference line); and (c) increase in spike
threshold (lack of spikes in black traces in all rows except the control, red
and green traces in row iv). (B) The amplitude of the hump evoked by a
constant test stimulus (600 pA, 5 ms) decreases as the amplitude of the
response evoked by the conditioning stimulus increases (color-coded, 3 ms).
The maximum effect is caused when a conditioning spike is evoked. Note the
undershoot after the sub-threshold stimuli (arrows).
|
|
View larger version (23K):
[in a new window]
|
Fig. 10. Post-spike increase in membrane conductance as revealed by passing steady
current. (A) Single spikes evoked by a brief pulse (bars) while passing
different intensities of steady current (black traces). Superimposed gray
traces correspond to control spikes without passing current. Note the
inversion of the after-hyperpolarization when the basal membrane potential
surpassed –80 mV (horizontal broken line). (B) V/I
plots obtained in another cell, before (open symbols), 5 ms after (gray
symbols) and 10 ms after the spike (black symbols). The linear relationship 5
ms after the spike (gray symbols) indicates that the spike causes an increase
in conductance similar to that caused by the tonic depolarization of the
membrane 1 ms before the spike (open symbols).
|
|
View larger version (54K):
[in a new window]
|
Fig. 11. The refractory period matches the duration of the low-responsiveness
window. Averaged spherical cell responses obtained from 10 series of current
test steps of constant intensity applied at different delays after a
conditioning spike (traces) are superimposed on the amplitude of the
magnocellularis nucleus responses to the conspecific-generated EOD shown in
Fig. 2, normalized as a
percentage of the amplitude of the self-generated EOD (black circles). (A)
In vitro test step at 400 pA, conspecific located antiparallel at 6
cm; (B) in vitro test step 430 pA, conspecific located antiparallel
at 3 cm. Resting membrane potential –70 mV, conditioning step 500 pA. In
order to compare both graphs, zero magnocellularis response was aligned with
the critical depolarization level and 100% magnocellularis response with the
largest spike provoked by the self-generated EOD (right axis).
|
|
© The Company of Biologists Ltd 2006
