Spike Timing-Dependent Plasticity in CA1 Pyramidal Neuron-Controlling Hippocampal Circuits: a Model Study
Spike timing-dependent plasticity (STDP) plays an important role in sculpting neural circuits to store information in the hippocampus, since motor learning and memory are thought to be closely linked with this type of synaptic plasticity. We built a computational model to study the potential le...
Збережено в:
| Дата: | 2014 |
|---|---|
| Автори: | , , , |
| Формат: | Стаття |
| Мова: | English |
| Опубліковано: |
Інститут фізіології ім. О.О. Богомольця НАН України
2014
|
| Назва видання: | Нейрофизиология |
| Онлайн доступ: | http://dspace.nbuv.gov.ua/handle/123456789/148293 |
| Теги: |
Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
|
| Назва журналу: | Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| Цитувати: | Spike Timing-Dependent Plasticity in CA1 Pyramidal Neuron-Controlling Hippocampal Circuits: a Model Study / H. Ren, S.Q. Liu, X. Zhang, Ya. Zeng // Нейрофизиология. — 2014. — Т. 46, № 4. — С. 335-342. — Бібліогр.: 19 назв. — англ. |
Репозитарії
Digital Library of Periodicals of National Academy of Sciences of Ukraine| id |
irk-123456789-148293 |
|---|---|
| record_format |
dspace |
| spelling |
irk-123456789-1482932019-02-18T01:25:59Z Spike Timing-Dependent Plasticity in CA1 Pyramidal Neuron-Controlling Hippocampal Circuits: a Model Study Ren, H. Liu, S.Q. Zhang, X. Zeng, Ya. Spike timing-dependent plasticity (STDP) plays an important role in sculpting neural circuits to store information in the hippocampus, since motor learning and memory are thought to be closely linked with this type of synaptic plasticity. We built a computational model to study the potential learning rule by linearly changing the synaptic weight and number of the synapses involved. The main findings are the following: (i) changes in the synaptic weight and number of synapses can lead to different long-term changes in the synaptic efficacy; (ii) the first spike pair of two neurons exerts a great influence on the subsequent spike pair; a pre-post spiking pair reinforces the subsequent paired spiking, while a post-pre spiking pair depresses this paired spiking; (iii) when the synaptic weight and synaptic number change, the interval in the first spiking pair is reduced, which directly influences the first spiking pair, and (iv) when a stellate neuron is stimulated weakly or the capacitance of a CA1 pyramidal neuron is decreased, LTP is produced more easily than LTD; in the opposite case, LTD is produced more readily; an increase of the synaptic number can promote activation of CA1 pyramidal neurons. Пластичність, залежна від часу генерації імпульсів (spike timing-dependent plasticity – STDP), відіграє важливу роль у функціонуванні нейронних мереж гіпокампа, що накопичують інформацію; вважають, що моторне навчання та пам’ять тісно пов’язані зі синаптичною пластичністю саме цього типу. Ми створили комп’ютерну модель, щоб вивчити можливі закономірності в процесі навчання, залежні від лінійних змін синаптичної ваги та кількості залучених синапсів у таких мережах. Основні експериментальні знахідки були наступними: 1) варіювання синаптичної ваги та числа синапсів можуть призводити до різних тривалих змін ефективності синаптичної передачі; 2) перша пара імпульсів, генерована двома синаптично пов’язаними нейронами, здійснює потужний вплив на наступну пару імпульсів; пара імпульсів у послідовності “пре–пост” полегшує генерацію наступної пари імпульсів, тоді як пара імпульсів у послідовності “пост–пре” пригнічує таку парну генерацію; 3) коли змінюються синаптична вага та кількість залучених синапсів, міжімпульсний інтервал у першій парі скорочується, тобто реалізується прямий вплив на характеристики такої пари; 4) коли інтенсивність стимуляції зірчастого нейрона є низькою або ємність мембрани пірамідного нейрона зони CA1 є зменшеною, тривала потенціація синаптичної передачі індукується легше, ніж тривала депресія; в іншому випадку легше виникає тривала депресія; збільшення кількості синапсів сприяє активації пірамідних нейронів у зоні CA1. 2014 Article Spike Timing-Dependent Plasticity in CA1 Pyramidal Neuron-Controlling Hippocampal Circuits: a Model Study / H. Ren, S.Q. Liu, X. Zhang, Ya. Zeng // Нейрофизиология. — 2014. — Т. 46, № 4. — С. 335-342. — Бібліогр.: 19 назв. — англ. 0028-2561 http://dspace.nbuv.gov.ua/handle/123456789/148293 519.876.2+612.014 en Нейрофизиология Інститут фізіології ім. О.О. Богомольця НАН України |
| institution |
Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| collection |
DSpace DC |
| language |
English |
| description |
Spike timing-dependent plasticity (STDP) plays an important role in sculpting neural circuits
to store information in the hippocampus, since motor learning and memory are thought to be
closely linked with this type of synaptic plasticity. We built a computational model to study
the potential learning rule by linearly changing the synaptic weight and number of the synapses involved. The main findings are the following: (i) changes in the synaptic weight and number of synapses can lead to different long-term changes in the synaptic efficacy; (ii) the first
spike pair of two neurons exerts a great influence on the subsequent spike pair; a pre-post spiking pair reinforces the subsequent paired spiking, while a post-pre spiking pair depresses this
paired spiking; (iii) when the synaptic weight and synaptic number change, the interval in the
first spiking pair is reduced, which directly influences the first spiking pair, and (iv) when a
stellate neuron is stimulated weakly or the capacitance of a CA1 pyramidal neuron is decreased,
LTP is produced more easily than LTD; in the opposite case, LTD is produced more readily; an
increase of the synaptic number can promote activation of CA1 pyramidal neurons. |
| format |
Article |
| author |
Ren, H. Liu, S.Q. Zhang, X. Zeng, Ya. |
| spellingShingle |
Ren, H. Liu, S.Q. Zhang, X. Zeng, Ya. Spike Timing-Dependent Plasticity in CA1 Pyramidal Neuron-Controlling Hippocampal Circuits: a Model Study Нейрофизиология |
| author_facet |
Ren, H. Liu, S.Q. Zhang, X. Zeng, Ya. |
| author_sort |
Ren, H. |
| title |
Spike Timing-Dependent Plasticity in CA1 Pyramidal Neuron-Controlling Hippocampal Circuits: a Model Study |
| title_short |
Spike Timing-Dependent Plasticity in CA1 Pyramidal Neuron-Controlling Hippocampal Circuits: a Model Study |
| title_full |
Spike Timing-Dependent Plasticity in CA1 Pyramidal Neuron-Controlling Hippocampal Circuits: a Model Study |
| title_fullStr |
Spike Timing-Dependent Plasticity in CA1 Pyramidal Neuron-Controlling Hippocampal Circuits: a Model Study |
| title_full_unstemmed |
Spike Timing-Dependent Plasticity in CA1 Pyramidal Neuron-Controlling Hippocampal Circuits: a Model Study |
| title_sort |
spike timing-dependent plasticity in ca1 pyramidal neuron-controlling hippocampal circuits: a model study |
| publisher |
Інститут фізіології ім. О.О. Богомольця НАН України |
| publishDate |
2014 |
| url |
http://dspace.nbuv.gov.ua/handle/123456789/148293 |
| citation_txt |
Spike Timing-Dependent Plasticity in CA1 Pyramidal Neuron-Controlling Hippocampal Circuits: a Model Study / H. Ren, S.Q. Liu, X. Zhang, Ya. Zeng // Нейрофизиология. — 2014. — Т. 46, № 4. — С. 335-342. — Бібліогр.: 19 назв. — англ. |
| series |
Нейрофизиология |
| work_keys_str_mv |
AT renh spiketimingdependentplasticityinca1pyramidalneuroncontrollinghippocampalcircuitsamodelstudy AT liusq spiketimingdependentplasticityinca1pyramidalneuroncontrollinghippocampalcircuitsamodelstudy AT zhangx spiketimingdependentplasticityinca1pyramidalneuroncontrollinghippocampalcircuitsamodelstudy AT zengya spiketimingdependentplasticityinca1pyramidalneuroncontrollinghippocampalcircuitsamodelstudy |
| first_indexed |
2025-07-12T19:04:17Z |
| last_indexed |
2025-07-12T19:04:17Z |
| _version_ |
1837469073822187520 |
| fulltext |
NEUROPHYSIOLOGY / НЕЙРОФИЗИОЛОГИЯ.—2014.—T. 46, № 4 335
UDC 519.876.2+612.014
H. REN,1 S.Q. LIU,1 X. ZHANG,1 AND YA. ZENG2
SPIKE TIMING-DEPENDENT PLASTICITY IN CA1 PYRAMIDAL NEURON-
CONTROLLING HIPPOCAMPAL CIRCUITS: A MODEL STUDY
Received 21.09.13
Spike timing-dependent plasticity (STDP) plays an important role in sculpting neural circuits
to store information in the hippocampus, since motor learning and memory are thought to be
closely linked with this type of synaptic plasticity. We built a computational model to study
the potential learning rule by linearly changing the synaptic weight and number of the synap-
ses involved. The main findings are the following: (i) changes in the synaptic weight and num-
ber of synapses can lead to different long-term changes in the synaptic efficacy; (ii) the first
spike pair of two neurons exerts a great influence on the subsequent spike pair; a pre-post spik-
ing pair reinforces the subsequent paired spiking, while a post-pre spiking pair depresses this
paired spiking; (iii) when the synaptic weight and synaptic number change, the interval in the
first spiking pair is reduced, which directly influences the first spiking pair, and (iv) when a
stellate neuron is stimulated weakly or the capacitance of a CA1 pyramidal neuron is decreased,
LTP is produced more easily than LTD; in the opposite case, LTD is produced more readily; an
increase of the synaptic number can promote activation of CA1 pyramidal neurons.
KEYWORDS: computational model, hippocampus, spike timing-dependent plasticity
(STDP), synaptic weight and number.
1 South China University of Technology, Department of Mathematics,
Guangzhou, China.
2 Beijing University of Technology, Biomedical Engineering Center, Beijing,
China.
Correspondence should be addressed to Sh. Liu or Ya. Zeng
(e-mail: yjzeng@bipu.edu.cn).
INTRODUCTION
Since an observation of persistent enhancement of the
intensity of synaptic transmission in the hippocampus
induced by tetanic stimulation [1], i.e., a phenomenon
now generally called long-term potentiation (LTP),
studies of activity-dependent synaptic plasticity have
become one of the scientific frontier and hot topics
in neurobiology [2, 3]. Later studies have further
addressed the importance of the temporal order of
presynaptic and postsynaptic spiking for long-term
modifications (LTP and long-term depression, LTD)
of a variety of glutamatergic synapses and have
defined the “critical windows” for spike timing. When
presynaptic spiking precedes postsynaptic spiking
(hereafter referred to as “pre-post” within a window of
several tens of milliseconds, LTP is induced, whereas
spiking of the reverse order (“post-pre”) leads to
LTD. This form of activity-dependent LTP/LTD is
now referred to as spike timing-dependent plasticity
(STDP) [4].
Levy and Steward have examined the effect of
relative millisecond-level timing of presynaptic
and postsynaptic action potentials (APs) on the
plasticity [5]. A later work, by Bi and Poo [6],
underscored the importance of precise spike timing,
synaptic strength, and postsynaptic cell type in the
activity-induced modification of central synapses
and suggested that the Hebb’s rule may need to be
incorporated in a quantitative consideration of spike
timing taking into account narrow and asymmetric
windows for the induction of synaptic modification.
Recently, Works [7] continued to study STDP in in
vivo systems to reveal several layers of complexity in
STDP. According to a mathematical fitting method of
exponential function by Bi and Poo [6], Vassilis et al.
[8] built a computational model aimed at investigation
of biophysical mechanisms by which storage and
recall of spatio-temporal input patterns are achieved
in the CA1 microcircuitry.
On the base of a real connection system between
different cells in the hippocampus circuit and using
computational approaches, we studied the interaction
of different inputs, as well as the STDP learning
regularities and effects of changes in the parameters of
synaptic action, on this phenomenon. In addition, some
NEUROPHYSIOLOGY / НЕЙРОФИЗИОЛОГИЯ.—2014.—T. 46, № 4336
H. REN, S.Q. LIU, X. ZHANG, AND YA. ZENG
meticulous potentiation and depression phenomena
can be simultaneously observed.
METHODS
Taking into consideration results of the anatomical ex-
periments, the connection system between cells, and
physiological parameters of each cell, we tried to con-
struct a computational hippocampus network maximal-
ly similar to the real system where all three kinds of the
neurons of the rat hippocampus were presented. These
were a CA1 pyramidal neuron, a stellate neuron, and a
CA3 pyramidal neuron. In order to better simulate real
properties of this circuit, we have used only three typi-
cal neurons to build the circuit, where perforant-path fi-
bers project to the hippocampus from the entorhinal cor-
tex and form excitatory synapses with the apical dendrite
of the CA1 pyramidal neuron, while the CA3 pyramidal
neuron is connected with the CA1 pyramidal neuron; the
latter also forms excitatory synapses through a set of the
fibers called Schaffer collaterals [9] (Fig. 1).
According to the results of natural experiments, the
configuration of the hippocampus circuit is easy to be
understood, and the sequence of events in this system
is also very specific. The physiological parameters
for this model related to each neuron were described
in detail in the preceding papers [9, 10] and in the
Appendix. This allowed us to construct our model
hippocampus circuit rather similar to the real system.
According to theoretical analysis, a single cell can be
characterized as composed by a conduction model and
a compartment model. Different parts of the neuron can
be described by different compartment numbers. The
basic theory of electrical signal transmission is the Rall’s
cable model; its discrete format corresponds to a neuronal
compartment model [11-14]. We should notice that
different cell compartments possess different ion channels
and some initial ion variables. In general, the parameters
and neuronal equations can be presented as follows:
In these equations, C, x = x(V); Icompartment and Isyn
separately indicate the cell membrane capacitance,
open or closed channels, external stimulus, compart-
ment currents, and synaptic currents. The properties
of chemical synapses were used to describe the con-
nections between the presented cells, and these special
connections can be described by the respective excit-
atory or inhibitory actions [14].
The constructed network was realized within the
NEURON environment, and we used MATLAB
software to process the data. The simulated results
have been repeatedly verified.
F i g. 1. Scheme of the modeled hippocampal circuit. A CA3 pyramidal neuron, a CA1 pyramidal neuron, an EC stellate neuron, interconnecting
fibers, and sites of stimulation are shown.
Р и с. 1. Схема модельованої гіпокампальної нейронної мережі.
Schaffer collaterals
perforant path
stim 1
stim 2
CA3 pyramidal neuron CA1 pyramidal neuron EC stellate neuron
NEUROPHYSIOLOGY / НЕЙРОФИЗИОЛОГИЯ.—2014.—T. 46, № 4 337
SPIKE TIMING-DEPENDENT PLASTICITY IN CA1 PYRAMIDAL NEURON-CONTROLLING HIPPOCAMPAL CIRCUITS
RESULTS
The synaptic plasticity is ability of the synaptic
connection between two neurons to change its efficacy
in response to either use or disuse of the transmission
over synaptic pathways [15]. Here we simulated the
two situations:
1) The synaptic conductance of NMDA receptors
is the product of 0.035 and synaptic weight, while
the synaptic conductance of AMPA receptors is the
product of 0.001 and synaptic weight. Then, we
linearly changed the synaptic weight, which is the
same as a linearly change in the synaptic conductance.
2) Linearly change in the number of the synapses.
Here, we give a description of the term “normalized
EPSP slope”. Under initial conditions, the first spike
generated by the postsynaptic neuron is marked as
AP0. If we change the prerequisites, the postsynaptic
neuron will generate a new first AP marked as V1.
Considering this, we can characterize the “normalized
EPSP slope” by k.
k = (V1 – V0)/V0.
LTP and LTD Produced in the CA1 Pyramidal
Neuron by Perforant Path Synapses. Results of the
previous research [16] showed that direct sensory in-
formation arriving at distal CA1 synapses through the
perforant path provide compartmentalized instructive
signals assessing the saliency of mnemonic informa-
tion propagated through the hippocampal circuit to
proximal synapses. Here, we mostly analyze the im-
pact of perforant path synapses.
At the beginning, we only stimulated the stellate
cell through a constant input, and there was no input
from Schaffer collaterals to the CA1 pyramidal neu-
ron. When initially the synapse weight was small, the
system always worked in the pre-post firing pattern;
the critical window shows that the LTP phenomenon
is realized within the time borders shown in the ex-
periment [17]. The first spiking pair interval demon-
strated a tendency toward decrease when the synaptic
weight increases (Fig. 2A). If we increase the synaptic
weight to a certain large value, the system always be-
gins to work corresponding to the post-pre firing pat-
tern, where LTD can develop (C).
When the number of synapse is initially small, the
0.4
–0.06
–1.6 –1.2 –0.8 –0.4 0
–0.05
–0.04
–0.03
–0.02
–0.01
0
0.01
0.4
0.3
0.2
0.1
0 00 2 4 6 8 10 12 141 2 3 4 5 6 7 8 9 10 msec
msec msec
msec
0.5
0.5
0.6
0.6
0.7
0.7
0.8
0.8
0.9
0.9
1.0
1.0
1.1
1.2
1.3
1.4
%
% %C
A
D
B
%
F i g. 2. The critical window. Effects of changes in the perforant-path synaptic number and weight. A) Synaptic number is 60, synaptic
weight varies from 0.08 to 0.35. B) Synaptic weight is 0.5, synaptic number varies from 15 to 51. C) Synaptic number is 60, synaptic weight
varies from 1.45 to 0.4. D) Synaptic weight is 0.5, synaptic number varies from 81 to 54.
Р и с. 2. Вплив кількості синапсів та синаптичної ваги зв’язків перфорантного шляху на критичне вікно.
–0.040
–0.035
–0.030
–0.025
–0.020
–0.015
–0.010
–0.005
0
–1.0 –0.8 –0.6 –0.4 –0.2 0
NEUROPHYSIOLOGY / НЕЙРОФИЗИОЛОГИЯ.—2014.—T. 46, № 4338
H. REN, S.Q. LIU, X. ZHANG, and YA. ZENG
system also shows a pre-post firing pattern. If we in-
crease the synaptic number up to 51 (the system in this
case works in the pre-post firing mode), the critical
window shows that the LTP phenomenon is manifested
(Fig. 2B). On the contrast, when the synapse number
is decreased to 54 from a larger number, the LTD phe-
nomenon also can be reached (D).
During LTP, the first pair of spikes is interpreted as
playing a dominant role in the restriction of subsequent
postsynaptic spiking. The first pre-post spiking pair
can intensify the subsequent paired spiking; never-
theless, potentiation would recede when the firing in-
terval descends (Fig. 3 A, B). During LTD, however,
the inversed conclusion can be reached (C, D).
According to the computational simulation, both
changes of the synaptic weight and synaptic number
would lead to the replacement of LTP and LTD. But
how the two long-term changes provide storing and
delivering of information? This question is quite dif-
ficult to prove. Moreover, the computational results
can inspire us to test and simulate the various factors,
which is extremely complicate to be realized in the
biological experiments to complete the STDP theory.
LTP Produced in the CA1 Pyramidal Neuron
by Schaffer-Collateral Synapses. Clarke and Ste-
phen [18] showed that selective induction of differ-
ent forms of LTP is achieved via spatial segregation of
functionally distinct calcium signals, where activation
of NMDA receptors is necessary. So, we studied the
impact of the Schaffer-collateral synapses on the CA1
pyramidal neuron to further understand the phenome-
non of synaptic plasticity.
Here we also stimulated the CA3 pyramidal neuron
via a constant input. In the same way, we still studied
the impact of the Schaffer-collateral synaptic weight
and synaptic number. We found that only LTP was in-
duced by increases in the synaptic weight and synap-
tic number (Fig. 4). Both the effect of the first spiking
pair and the trend of the spiking interval were simi-
lar with the results described in the first paragraph
of Results. The difference was that we failed to find
the post-pre pattern of the first spiking pair interval,
F i g. 3. The first spiking pair restricts
subsequent postsynaptic spiking. The
perforant- path synaptic weight and number
are: A) top 0.2 and 36, bottom 0.5 and 36;
B) top 0.5 and 12, bottom 0.5 and 36; C)
top 0.7 and 96, bottom 0.2 and 96; D) top
0.8 and 96, bottom 0.8 and 48, respectively.
Red line shows the firing pattern of the
stellate neuron, and black line shows that
of the CA1 pyramidal neuron.
Р и с. 3. Вплив параметрів синапсів
перфорантного шляху, що визначають
характеристики першої пари імпульсів,
на генерацію другої пари імпульсів.
20 mV
EC neuron
CA1
10 msec
A B
C D
0
4 6 8 10 12 14 16msec
0.1
0.2
0.3
0.4
0.5
0.6
0.7
%
BA
20 mV
10 msec
F i g. 4. The critical window (A) and changes in the postsynaptic spiking (B) at changes in the parameters of synapses of Schaffer
collaterals. A) Synaptic number is 48, and synaptic weight varies from 0.2 to 3.0. B) Restriction of the following postsynaptic spiking by
the first spiking pair; blue line shows firing of the CA3 pyramidal neuron, and black line shows that of the CA1 pyramidal neuron. The
synaptic weight and number of synapses of Schaffer collaterals are: top, synaptic weight is 0.7 and number of synapses is 54; bottom, the
respective parameters are 1.4 and 54.
Р и с. 4. Вплив кількості синапсів та синаптичної ваги зв’язків колатералей Шафера на критичне вікно.
NEUROPHYSIOLOGY / НЕЙРОФИЗИОЛОГИЯ.—2014.—T. 46, № 4 339
SPIKE TIMING-DEPENDENT PLASTICITY IN CA1 PYRAMIDAL NEURON-CONTROLLING HIPPOCAMPAL CIRCUITS
which can induce LTD.
From the aforementioned research, we acquired
some properties of LTP and LTD; now we will sum-
marize their difference and similarity. The difference
mainly is related to the initial state, magnitude, and ef-
fect of the first spiking pair interval (detailed descrip-
tion was given in the above paragraph). The similarity
is that the magnitude of the effect of the first spiking
pair interval demonstrated a decreasing trend in both
LTP and LTD. Transformation of the role of the first
spiking pair depends on the synaptic weight and num-
ber of synapses. The smaller the synaptic weight and
number of synapses, the more obvious the inhibitory
function. We should, however, notice that the first fir-
ing interval also depends on the synaptic weight and
number of synapses. We found that when the first spik-
ing pair interval is reduced, the peak voltages in the
CA1 pyramidal neuron decrease, and this mode of in-
hibition depends on firing of the presynaptic stellate
neuron. The less pre-bursting spikes, the more obvious
the inhibitory function (Fig. 5).
Factors that can Affect the Pattern of the First
Spiking. Based on the former analyses, we tried to
research the factors that can influence the firing pat-
tern of the first spiking. When, initially, the synap-
tic weight is large, LTD is induced by decrease in the
synaptic weight. On the contrast, when the synaptic
weight is initially small, LTP is induced by increase in
this weight. We should, however, pay special attention
to the critical value in relations between LTD and LTP.
From the simulated graph of the relationship be-
tween the critical value and intensity of an external
stimulus, it is easy to find that when the external stim-
ulus intensity increases, the critical value will de-
crease. Then we continued to study the relationship
between the critical value and membrane capacitance
of the CA1 pyramidal neuron. It was found that when
the membrane capacitance increases, the critical val-
ue will increase. In other words, LTP was easier to be
induced than LTD with a larger membrane capacitan-
0
–0.1
–40
–30
–20
–10
0
10
20
30
–0.2
0.2
0.4
0.6
0.8
0
msec msec–1 –1
% %
0 1 2 3 4 5 6 7 8
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 1 2 3 4 5
–1 –1 –10 0 01 1 12 2 23 3 34 4 4
A B
C
F i g. 5. Comparison of the influences of the first spiking interval on induction of LTD and LTP (A and B) and the relationship between the
peak voltage in the CA1 pyramidal neuron and the first spiking interval at different firing patterns of the presynaptic stellate neuron (C).
Р и с. 5. Порівняльний вплив інтервалу між імпульсами першої пари на індукцію тривалої депресії або тривалої потенціації
(А, В) та залежність зміни потенціалу у пірамідному нейроні від першого інтервалу при різних патернах розряду в зірчастому
нейроні (С).
NEUROPHYSIOLOGY / НЕЙРОФИЗИОЛОГИЯ.—2014.—T. 46, № 4340
H. REN, S.Q. LIU, X. ZHANG, and YA. ZENG
ces of the CA1 pyramidal neuron or a weaker external
stimulus applied to this neuron (Fig. 6).
Function of Schaffer-Collateral Synapses on the
CA1 Pyramidal Neuron. Jarsky et al. [19] examined
the function of the distal synaptic inputs, and the
results led to three predictions that were confirmed in
the experiments using rat hippocampal slices. On the
basis of the cited research, we primarily studied the
function of Schaffer-collateral synapses on the CA1
pyramidal neuron.
Here, the CA1 pyramidal neuron is only connected
with the stellate neuron that receives a constant ex-
ternal stimulus. In this case, we can evaluate the rela-
tionship between the necessary synaptic number and
synaptic weight. To study how the Schaffer-collateral
inputs affect this relationship, we connected the CA3
pyramidal neuron with the CA1 pyramidal neuron once
again. The necessary synaptic number decreased in
both inputs in contrast to the situation without Schaf-
fer-collateral inputs. The Schaffer-collateral input can
better promote the efficiency of activation of the CA1
pyramidal neuron (Fig. 7A). Then, we fixed the syn-
aptic weights and could see that activation of the CA1
pyramidal neuron would be promoted by increasing
the numbers of two kinds of synapses (B).
Then, we analyzed the influence of Schaffer
collaterals on the perforant path by changing the
perforant-path synaptic weight, and LTD and LTP were
reproduced again. Nevertheless, the range of k is very
small in the case of LTD, and the change of k is wider
in the case of LTP than those without application of
the stimulus to Schaffer collaterals (Fig. 8 A, B).
Based on the role of the first spiking pair, we have
discussed above, when the first spiking pair has a pre-
post firing pattern, the subsequent presynapse spiking
can promote firing of the postsynapse in the perforant-
path synapses. However, such promotion decreases
after firing of the CA3 pyramidal neurons even
though LTP is expected to be induced in the Schaffer-
collateral synapses (Fig. 8C).
0.30
0 0
00
0 0.2 0.4 0.6 0.8 1
0.2
0.1
0.5
0.4
0.04 0.08
0.3
0.2
0.4
0.6
0.8
1
20
40
60
80
100
120
140
0.35
0.5 0.501.0
pre-post
post-pre post-pre
pre-post
1.01.5 1.52.0 2.02.5 2.53.0 3.03.5 3.54.0 4.0
0.32
0.40
0.34
0.450.36
0.50
0.38
0.55
0.40
0.42
B
B
A
A
F i g. 6. Relationship between the critical value at
two patterns and the intensity of an external stimulus
(A) and relationship between the critical value at two
patterns and the membrane capacitance of the CA1
pyramidal neuron (B).
Р и с. 6. Взаємозалежність критичного значення
при двох патернах імпульсації, інтенсивності
зовнішнього стимулу (А) та ємності мембрани
пірамідного нейрона СА1 (В).
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10
20 mV
10 msec
F i g. 7. Relationship between the synaptic weight of perforant-path synapses and necessary synaptic number. Blue line is such relationship
at stimulation of only the perforant path; red line is that at stimulation of both the perforant path and Schaffer-collaterals (A). Activation
of the CA1 pyramidal neuron is promoted because increases of the numbers of synapses of the above-mentioned types. Blue points show
situations where the CA1 pyramidal neuron was not activated, and red points show those where this neuron was successfully excited.
Р и с. 7. Залежність між синаптичною вагою зв’язків перфорантного шляху та колатералей Шафера і необхідною кількістю
синаптичних контактів, котра визначає активацію пірамідного нейрона зони CA1.
NEUROPHYSIOLOGY / НЕЙРОФИЗИОЛОГИЯ.—2014.—T. 46, № 4 341
SPIKE TIMING-DEPENDENT PLASTICITY IN CA1 PYRAMIDAL NEURON-CONTROLLING HIPPOCAMPAL CIRCUITS
DISCUSSION
Thus, we tried to study the mechanisms of the
hippocampus circuit model using variations of the
firing pattern qualities. We mainly studied interaction
of the two synapse groups on the CA1 pyramidal
neuron and examined the synaptic plasticity. There are
several important points to be discussed.
1. Changes in the synaptic weight and synaptic
number between the CA1 pyramidal neuron and stellate
neuron will lead to long-term changes in the former
neuron, namely manifestations of LTD and LTP. The
first spike pair generated by the postsynaptic neuron
(CA1 pyramidal neuron) plays an important role in
modification of the subsequent spike pair. The pre-
post firing pattern plays a positive role in facilitation
of the subsequent spiking. On the contrary, the post-
pre firing pattern plays a negative role.
2. Changes in the synaptic weight and synaptic
number between the CA1 and CA3 pyramidal neurons
only lead exclusively to LTP in the CA1 pyramidal
neuron. When we change the synaptic weight, the first
spiking pair interval will be reduced, which directly
inhibit the first spiking pair changes.
3. When we apply a weak external stimulus to
the neurons or decrease the capacitance of the CA1
pyramidal neuron, LTP is produced more easily than
LTD. In addition, Schaffer collaterals can better
promote activation of the CA1 pyramidal neuron;
however, they inhibit the potentiation of the first
spiking pair.
Based on previous analyses, the constructed
hippocampal network has a clear network configuration
and mimics a number of functions. In this paper, we
mainly discussed the synapse modification due to
linear changes in the synaptic number and weight
and interaction between the perforant-path inputs and
Shaffer-collateral inputs. In fact, some results of our
study are very similar to those obtained in realistic
experiments. However, it is difficult to formulate
general learning rules in biological experiments; thus,
model computation results will be helpful with respect
to future research.
0 0
–0.1
a1 b1 c1
a2 b2 c2
0.1 0.05
0.2 0.10
0.3 0.15
0.4 0.20
0.5 0.25
0.6 0.30
0.7
% %
0.35
msec msec–1 0 2 4 6 8 100 1 2 3 4 5 6 7 8
B
C
A
40 mV
20 msec
F i g. 8. The same as in Fig. 7, but at stimulation of the perforant path; black and blue dots show the cases of LTP and LTD, respectively (A);
effects of combined stimulation of synapses of the perforant path and Schaffer collaterals (B); illustration of the firing patterns of different
neurons (C; blue, red, and black lines are those of the CA3 pyramidal neuron, EC stellate neuron, and CA1 pyramidal neuron, respectively).
Р и с. 8. Залежність індукції тривалої потенціації або тривалої депресії від сили стимуляції перфорантного шляху та колатералей
Шафера.
NEUROPHYSIOLOGY / НЕЙРОФИЗИОЛОГИЯ.—2014.—T. 46, № 4342
H. REN, S.Q. LIU, X. ZHANG, and YA. ZENG
Appendix. Supporting information
Supplementary data associated with this article are
presented in the online version at http://neuromorpho.org/ and
http://senselab.med.yale.edu/modeldb/
Х. Рен1, Ш. К´ю. Лью1, Кс. Жанг1, Я. Зенг2
ЗАЛЕЖНА ВІД ЧАСУ ГЕНЕРАЦІЇ ІМПУЛЬСІВ
ПЛАСТИЧНІСТЬ У НЕЙРОННИХ МЕРЕЖАХ
ГІПОКАМПА, ЩО КОНТРОЛЮЮТЬ АКТИВНІСТЬ
ПІРАМІДНИХ НЕЙРОНІВ ЗОНИ CA1: МОДЕЛЬНЕ
ДОСЛІДЖЕННЯ
1 Південнокитайський технологічний університет, Гуан-
чжоу (Китай).
2 Центр біомедичної інженерії Пекінського технологічного
університету (Китай).
Р е з ю м е
Пластичність, залежна від часу генерації імпульсів (spike
timing-dependent plasticity – STDP), відіграє важливу роль у
функціонуванні нейронних мереж гіпокампа, що накопичу-
ють інформацію; вважають, що моторне навчання та пам’ять
тісно пов’язані зі синаптичною пластичністю саме цього
типу. Ми створили комп’ютерну модель, щоб вивчити мож-
ливі закономірності в процесі навчання, залежні від ліній-
них змін синаптичної ваги та кількості залучених синапсів
у таких мережах. Основні експериментальні знахідки були
наступними: 1) варіювання синаптичної ваги та числа си-
напсів можуть призводити до різних тривалих змін ефектив-
ності синаптичної передачі; 2) перша пара імпульсів, гене-
рована двома синаптично пов’язаними нейронами, здійснює
потужний вплив на наступну пару імпульсів; пара імпуль-
сів у послідовності “пре–пост” полегшує генерацію наступ-
ної пари імпульсів, тоді як пара імпульсів у послідовності
“пост–пре” пригнічує таку парну генерацію; 3) коли змі-
нюються синаптична вага та кількість залучених синапсів,
міжімпульсний інтервал у першій парі скорочується, тобто
реалізується прямий вплив на характеристики такої пари;
4) коли інтенсивність стимуляції зірчастого нейрона є низь-
кою або ємність мембрани пірамідного нейрона зони CA1 є
зменшеною, тривала потенціація синаптичної передачі інду-
кується легше, ніж тривала депресія; в іншому випадку лег-
ше виникає тривала депресія; збільшення кількості синапсів
сприяє активації пірамідних нейронів у зоні CA1.
REFERENCES
1. T. V. Bliss and T. Lomo, “Long-lasting potentiation of synaptic
transmission in the dentate area of the anaesthetized rabbit
following stimulation of the perforant path,” J. Physiol., 232,
331-356 (1973).
2. R. C. Malenka and S. A. Siegelbaum, Synaptic Plasticity, Johns
Hopkins Univ. Press (2001).
3. S. J. Martin, P. D. Grimwood, and R. G. Morris, “Synaptic
plasticity and memory: an evaluation of the hypothesis,” Annu.
Rev. Neurosci., 23, 649-711 (2000).
4. Y. Dan and M. M. Poo, “Spike timing-dependent plasticity:
from synapse to perception,” Physiol. Rev., 86, 1033-1048
(2006).
5. W. B. Levy and O. Steward, “Temporal contiguity requirements
for long-term associative potentiation/ depression in the
hippocampus,” Neuroscience, 8, No. 4, 791-797 (1983).
6. G. Q. Bi and M. M. Poo, “Synaptic modifications in cultured
hippocampal neurons: Dependence on spike timing, synaptic
strength, and postsynaptic cell type,” J. Neurosci., 18, No. 24,
10464-10472 (1998).
7. C. Natalia and Y. Dan, “Spike timing-dependent plasticity:
A hebbian learning rule,” Ann. Rev. Neurosci., 31, 25-46
(2008).
8. C. Vassilis, C. Stuart, and P. G. Bruce, “Encoding and
retrieval in a model of the hippocampal CA1 microcircuit,”
Hippocampus, 20, 423-446 (2010).
9. M. Royeck, M. T. Horstmann, S. Remy, et al., “Role of axonal
NaV1.6 sodium channels in action potential initiation of CA1
pyramidal neurons,” J. Neurophysiol., 100, No. 4, 2361-2380
(2008).
10. H. Peter, E. Daniel, B. Angela, et al., “Distinct classes of
pyramidal cells exhibit mutually exclusive firing patterns in
hippocampal area CA3,” Hippocampus, 18, No. 4, 411-424
(2008).
11. W. M. Yamada, C. Koch, and P. R. Adams, Multiple Channels
and Calcium Dynamics, MIT Press, Cambridge (1987),
pp. 97-134
12. M. Migliore and G. M. Shepherd, “Dendritic action potentials
connect distributed dendrodendritic microcircuits,” J. Comput.
Neurosci., 24, 207-221 (2008).
13. R. Michel, M. T. Horstmann, R. Stefan, et al., “Role of axonal
Na V 1.6 sodium channels in action potential initiation of CA1
pyramidal neurons,” J. Physiol., 4, 2361-2380 (2008).
14. L. Wang and S. Q. Liu, “Neural circuit and its functional roles
in cerebellar cortex,” Neurosci. Bull., 27, No. 3, 173-184
(2011).
15. J. R. Hughes, “Post-tetanic potentiation,” Physiol. Rev., 38,
No. 1, 91-113 (1958).
16. T. D. Joshua, T. David, and A. S. Steven, “A role for synaptic
inputs at distal dendrites: Instructive signals for hippocampal
long-term plasticity,” Neuron, 56, 866-879 (2007).
17. J. Tim, R. Alex, L. K. William, and S. Nelson, “Conditional
dendritic spike propagation following distal synaptic activation
of hippocampal CA1 pyramidal neurons,” Nat. Neurosci., 8,
1667-1676 (2005).
18. R. R. Clarke and J. R. Stephen, “Spatial segregation of
neuronal calcium signals encodes different forms of LTP in rat
hippocampus,” J. Physiol., 570, No. 1, 97-111 (2006).
19. T. Jarsky, A. Roxin, W. L. Kath, and N. Spruston, “Conditional
dendritic spike propagation following distal synaptic activation
of hippocampal CA1 pyramidal neurons,” Nat. Neurosci., 8,
No. 12, 1667-1676 (2005).
|