Biomedical Engineering Reference
In-Depth Information
Figure 7.7
Each spontaneous spike triggers the application of an electrical stimulus pulse to
the neural stalk, which initiates an antidromic spike in the axon of every supraop-
tic neuron, since all supraoptic neurons project to the pituitary via the neural stalk.
The antidromic spike evoked in the axon of the recorded neuron is extinguished by
collision but other antidromic spikes persist to invade the cell bodies of most neigh-
bouring neurons, and thence activates intranuclear connections and dendritic release
of oxytocin. [19]. (B) Every interspike interval (
t
1
) in a selected recording period
was paired with its predecessor (
t
2
) and preceding intervals
t
3
,
t
4
, ···
. to study the
dependence of current activity upon preceding activity-B shows the analysis of a
representative oxytocin neuron, the mean
t
1
(
standard error) is plotted against
t
2
before (left upper panel) and after (right upper panel) CCS, and the corresponding
interspike interval distributions before and after CCS (bottom panel). CCS stimula-
tion induces an increase in the proportion of short intervals, as seen in the interspike
interval histograms, and an increase in clustered firing, as shown by the positive re-
lationship between
t
1
and
t
2
. (C) Example of the mean
t
1
(
±
±
standard error) against
t
2
,or
t
2
+
t
5
(with linear or polynomial trend lines) for a represen-
tative neuron in control conditions (upper panels), during CCS (middle panels), and
during CCS + thapsigargin (bottom panels). From the top left panel in C it seems
there is little influence of
t
2
upon
t
1
, there is a negative regression, but with a very
shallow slope. However, this weak influence is long-lasting and so summates with
successive spikes, because in looking at
t
1
vs.
t
2
t
3
,or
t
2
+
t
3
+
t
4
+
t
5
, there is now a strong
inverse correlation. During CCS this negative relationship is still present but is pre-
ceded by a positive relationship, indicating a short-lasting positive feedback action
superimposed upon the normal slow negative feedback.
+
···
+
These results demonstrate structure in sequences of interspike intervals, and from its
characteristics we may conclude it to be the effect of the AHP.
Thus sufficient information seems to be available from the characteristics of spon-
taneous discharge activity to produce concise computational models that can mimic
this behavior closely when they incorporate features that appropriately describe the
impact of intrinsic, activity-dependent mechanisms. Such models are unique de-
scriptors of a particular neuronal phenotype, and are by design well-matched to ex-
perimental data, but are also capable of generating fresh insight into cell properties,
testing the coherence and feasibility of biological hypotheses, and capable of gener-
ating novel and counter-intuitive predictions.
We have concentrated on demonstrating this approach for an example neuron with
limited network connectivity. The oxytocin cell is an output neuron, with few axonal
collaterals to make any recurrent connection with other neurons in the CNS, hence
activity-dependent influences in activity primarily reflect intrinsic cell properties in
normal circumstances. However, this approach is potentially particularly appropri-
ate for analysing the behavior of neurons where activity-dependent influences are
mediated by interactions with other neurons. As far as the analytical approach is
concerned, this is indifferent to whether activity-dependent influences reflect intrin-
Search WWH ::
Custom Search