I N N O V A T I O N S
A N D
I D E A S
BASIC PRINCIPLES OF
SYNAPTIC PHYSIOLOGY ILLUSTRATED
BY A COMPUTER MODEL
Michael J. Davis
Department of Medical Physiology, Texas A&M University System Health Science Center,
College Station, Texas 77843
computer model is described that simulates many basic aspects of chemical
synapse physiology. The model consists of two displays, the first being a
pictorial diagram of the anatomical connections between two presynaptic
neurons and one postsynaptic neuron. Either or both of the presynaptic cells can be
stimulated from a control panel with variable control of the number of pulses and
firing rate; the resulting presynaptic action potentials are animated. The second
display plots the membrane potential of the postsynaptic cell versus time following
presynaptic stimulation. The model accurately simulates temporal and spatial summation when the presynaptic cells are arranged and stimulated in parallel and
simulates presynaptic inhibition when they are arranged and stimulated in series.
Excitatory and inhibitory postsynaptic potentials can be demonstrated by altering the
nature of the ionic conductance change occurring on the postsynaptic cell. The
effects on summation of changing length constant or time constant of the postsynaptic cell can also be illustrated. The model is useful for teaching these concepts to
medical, graduate, or undergraduate students and can also be used as a self-directed
computer laboratory exercise. It is available for free download from the internet.
A
ADV PHYSIOL EDUC 25: 1–12, 2001.
Key words: excitatory postsynaptic potential; inhibitory postsynaptic potential; time
constant; length constant; accommodation; temporal summation; spatial summation
ical and graduate students. Additionally, it needed to
demonstrate realistic responses of neuron-to-neuron
synapses, including temporal and spatial summation,
excitatory and inhibitory potentials, and the influences of the membrane time constant and length
(space) constant on the degree of summation.
The transmission of information across a chemical
synapse is a basic concept taught in most physiology
courses (2, 3). The usual method of teaching this
topic involves a sequential description of pre- and
postsynaptic events using static illustrations to indicate how several factors interact to determine the
summation of postsynaptic potentials. Laboratory
demonstrations of these processes are difficult and
expensive as they require live neurons impaled with
glass microelectrodes. My objective was to develop a
computer model to simulate many basic properties of
the chemical synapse. The model needed to be appropriate for a laboratory exercise for first-year med-
This article descibes a computer model that fulfills
these criteria. Surprisingly, it is also effective as a
dynamic classroom teaching aid. Although several
other computer models on synaptic physiology are
commercially available, this model is available in
Macintosh and Windows versions for free distribution
1043 - 4046 / 01 – $5.00 – COPYRIGHT © 2001 THE AMERICAN PHYSIOLOGICAL SOCIETY
VOLUME 25 : NUMBER 1 – ADVANCES IN PHYSIOLOGY EDUCATION – MARCH 2001
1
I N N O V A T I O N S
over the internet to interested faculty and students.
Also available is a detailed worksheet that can be used
as a self-directed laboratory exercise.
A N D
I D E A S
Two display windows appear on the screen (Fig. 1)
below the menu bar. On the left is a pictorial diagram
of the anatomical connections between two presynaptic neurons (cells A and B) and one postsynaptic
neuron (cell C). A diagram of a recording electrode
placed near the axon hillock of cell C indicates the site
of membrane potential (Em) measurements. Below
this display are located various controls to alter the
stimulus parameters, the anatomical arrangement between the cells, and membrane properties of the
postsynaptic cell. The display on the right shows the
change in Em (⌬Em) of cell C as a function of time.
METHODS
Software. All components of the model were developed using LabView (version 4.0, National Instruments, Austin, TX). The source code was subsequently complied as run-time (stand-alone) code
capable of executing on computers without the LabView Development System. The program consists of a
single “virtual instrument.” Testing and development
were performed on a 300 MHz Macintosh PowerPC
computer and a 233 MHz Dell computer. The model
will also run on slower hardware supporting either
platform. A control has been added to enable adjustment of the drawing time (1 ⫽ fast, 100 ⫽ slow) so
that the animation speed can be matched to the
power of the computer being used.
The default settings for the model are set to stimulate
one presynaptic cell (cell A) with one pulse at the
lowest (firing) rate. Cell A briefly flashes red when the
model starts up, or when the run button is pressed,
and the Em display shows the associated excitatory
postsynaptic potential (EPSP) on cell C, which has an
amplitude of 2 mV and duration of 2 ms. The timing
of the presynaptic stimulus pulses is also marked on
the Em versus time graph by gray pulses just below
the Em trace.
The layout of controls and graphs is optimized for
display on a color monitor supporting 832 ⫻ 624 or
800 ⫻ 600 resolution. The source code occupies
⬃450 kb of disk space, whereas the compiled code
with embedded libraries occupies ⬃2.5 Mb of disk
space and requires 6 Mb of system memory to execute. Macintosh and Windows versions, along with
the worksheet found in the APPENDIX, can be downloaded from the following internet address: (http//
mphywww.tamu.edu/davis/models/synapse.html.)
Calculation of postsynaptic Em. In the absence of
presynaptic stimulation, the resting potential of the
postsynaptic cell is stable at ⫺70 mV. Subthreshold
synaptic potentials, i.e., EPSPs, inhibitory postsynaptic potentials (IPSPs), or any combination of the two,
are seen as time-dependent deviations from the resting potential. Depolarizing stimuli that summate to
exceed the postsynaptic cell threshold of ⫺60 mV
result in an action potential on the postsynaptic cell.
Displays. After the model has loaded, a menu bar at
the top of the screen controls program execution.
The model initially loads, runs once, then stops. The
“run” button, an arrow-shaped button in the upper
left corner, is used to run the model again. In typical
use, the stimulus controls are adjusted with the model
off, then the run button is pressed and the displays are
updated. The model can also be started with the
button adjacent to the run button. This “continuous
run” mode repeatedly executes the program, allowing
changes to be made to any of the parameters “on-thefly.” The rate at which the displays are updated in this
mode can be adjusted using the “display speed” control in the lower right corner. The model can be
halted at any time using the “stop” button in the
menu.
The program logic flow is illustrated by the diagram in
Fig. 2. After presynaptic stimulation, the time course
of the ⌬Em is approximated using the equation:
⌬Em ⫽ a ⫻ t ⫻ exp(⫺a ⫻ t), where a is the time
constant and t is time. This parameter is calculated for
both presynaptic inputs. Their sum is computed at
each point in time for each of 50 time intervals and
added to the value of the resting Em. The degree of
summation of the two individual postsynaptic events
is determined by, and proportional to, the value of the
length constant for cell C, where 1 ⫽ complete summation and 0 ⫽ no summation. If the postsynaptic
cell Em reaches its threshold value of ⫺60 mV, the
explosive voltage change that follows is computed by
VOLUME 25 : NUMBER 1 – ADVANCES IN PHYSIOLOGY EDUCATION – MARCH 2001
2
I N N O V A T I O N S
A N D
I D E A S
FIG. 1.
The main display window of the model is shown. An animated view of pre- and postsynaptic cell firing is
at the top left. Bottom left: controls for stimulation of the presynaptic cells and for changing the properties
of the postsynaptic cell membrane. The graph to the right shows the change in membrane potential (Em) at
the axon hillock of cell C. In this example, a single action potential in cell A is indicated by the darker
appearance of axon A, and the resulting EPSP on cell C is indicated by the 2-mV depolarization shown on the
Em-vs.-time graph.
RESULTS
adding to Em a stored array of values generated from
Hodgkin-Huxley-style equations (1) used in a separate
model of the neuronal action potential (unpublished
observations).
The model is capable of simulating many aspects of
chemical synaptic physiology, including temporal and
spatial summation, presynaptic inhibition, the influence of different postsynaptic receptor ionic conductances, and the effect of changing postsynaptic membrane time constant or length constant.
Animation. To animate the sequence of pre- and
postsynaptic cell firing, a simple LabView trick is employed. Separate layers of drawings, colored or clear, are
made for each of the neuronal objects using Photoshop
(Adobe Systems) and exported to LabView as PICT
objects. These are imported into LabView and aligned
on the front panel display as elements of individual
cluster controls for which attribute nodes are created.
Each of the nodes is then controlled programmatically
by switching on or off its visibility attribute.
Temporal summation. Repeated stimulation of cell
A (with cell B silent) at the lowest firing rate (1)
results in multiple, successive, 2-mV EPSPs on cell C
that decay before any summation occurs. An example
using four pulses is shown in Fig. 3A. Because cell C
repolarizes completely before the next presynaptic
action potential arrives, it never reaches a value that is
VOLUME 25 : NUMBER 1 – ADVANCES IN PHYSIOLOGY EDUCATION – MARCH 2001
3
I N N O V A T I O N S
A N D
I D E A S
FIG. 2.
Flow diagram of the program logic. For details, see text. EPSP, excitatory postsynaptic
potential; AP, action potential.
more positive than ⫺68 mV. Under the same conditions, an increase in firing rate (to an arbitrary value of
7 in Fig. 3B) results in partial temporal summation of
the individual EPSPs on cell C, causing the Em of cell C
to depolarize to about ⫺66 mV. Further increases in
the firing rate or the number of pulses will produce
even more summation. If the model is placed in continuous run mode, it will continuously redraw and
recalculate these changes as the firing rate for cell A is
varied.
Spatial summation. To demonstrate spatial summation, it is first advisable to reset the controls, either by
manually returning the “# pulses” back to 1 or by
selecting the “reinitialize all to default” from the control menu at the top of the screen. If cell B is now set
VOLUME 25 : NUMBER 1 – ADVANCES IN PHYSIOLOGY EDUCATION – MARCH 2001
4
I N N O V A T I O N S
A N D
I D E A S
FIG. 3.
Demonstration of temporal summation when cell A is stimulated at a high firing rate (B) but not a low
firing rate (A). Cell B is silent. In each panel, the top halves of the controls have been repositioned near
the Em-vs.-time displays to conserve space.
FIG. 4.
Demonstration of spatial summation when cells A and B are stimulated at exactly the same point in time
(A) or when cell B is stimulated 0.6 ms after cell A (B). Spatial summation occurs in both instances but
is more easily visible in B.
VOLUME 25 : NUMBER 1 – ADVANCES IN PHYSIOLOGY EDUCATION – MARCH 2001
5
I N N O V A T I O N S
A N D
I D E A S
FIG. 5.
Effect of changing the nature of the postsynaptic receptor conductance change (i.e., receptor type). Same
conditions as in Fig. 1 except that the postsynaptic receptor conducts Kⴙ ions in this example, producing
an inhibitory postsynaptic potential (IPSP) rather than an EPSP (Fig. 1). This is shown by the darker trace
in A. Setting the conductance change to “K ⴙ Cl” (updated controller is not shown) results in a reduced
IPSP amplitude (lighter trace marked by arrow in A). B: when both cell A and cell B are stimulated with
a 1-ms delay between them, cell A causes an EPSP, whereas cell B causes an IPSP. The algebraic sum of
these events produces the biphasic waveform shown. Note: in A, the “remember previous trace” feature
was turned on to overlay the 2 traces.
to be stimulated simultaneously with cell A (using 1
pulse at the lowest firing rate), the two associated
EPSPs (2 mV each) summate completely to produce a
4-mV depolarization of cell C (Fig. 4A). It may not be
immediately obvious to students that spatial summation has occurred in this case as opposed an increase
in the amplitude of a single EPSP by some other
undefined mechanism. However, spatial summation
can be more clearly detected by adding a slight delay
to the pulse in cell B (to 1.6) in Fig. 4B; this shift is
indicated by a shift in the timing marks of the two
pulses below the Em trace. Now it becomes clear that
two individual postsynaptic events are summating.
Putting the model in continuous run mode allows the
degree of spatial summation to be controlled on-thefly by alterations in the value of “pulse delay.”
By controlling the number of pulses and the respective firing rates of cells A and B, virtually any combination of temporal and spatial summation can be
demonstrated. This is an effect that is very difficult to
demonstrate or appreciate using a static figure.
Changes in postsynaptic receptor conductance.
Whether presynaptic stimulation results in an EPSP or
IPSP on the postsynaptic cell depends on the nature
of the postsynaptic receptor. So-called “excitatory
transmitters” (a term that is actually a misnomer)
typically interact with postsynaptic receptors, ligandgated ion channels, that result in equal increases in
postsynaptic cell Na⫹ and K⫹ conductance (3). The
nicotinic ACh receptor is one example. This is the
default setting of the model and produces the EPSP
VOLUME 25 : NUMBER 1 – ADVANCES IN PHYSIOLOGY EDUCATION – MARCH 2001
6
I N N O V A T I O N S
A N D
I D E A S
if the equilibrium potential for Cl⫺ is equal to the
resting Em (⫺70 mV), as Cl⫺ movement will act to
anchor Em at its resting value and partially offset the
hyperpolarizing effect of K⫹ movement through the
same channel. Postsynaptic conductance can be
switched between a number of different settings:
“Na,” “Na ⫹ K,” K, K ⫹ Cl, or “Cl,” each of which
produces an appropriate change in the relative magnitude and sign of the resulting postsynaptic potential.
shown in Fig. 1. If the conductance of the postsynaptic membrane at the termination of cell A is set to “K,”
the Em of cell C will move toward the K⫹ equilibrium
potential (EK) when cell A is stimulated. Because EK is
normally more negative than the resting Em of cell C,
the result is an IPSP, as illustrated by the ⫺2-mV IPSP
in Fig. 5A. If the receptor conductance is set to “K ⫹
Cl,” simulating the action of a transmitter that produces an equal increase in postsynaptic K⫹ and Cl⫺
conductance, the amplitude of the resulting IPSP is
reduced from ⫺2 mV to approximately ⫺1 mV (arrow in Fig. 5A). To allow comparisons between these
two parameter sets, the button “remember previous
trace” needs to be set to “on” before the K ⫹ Cl
conductance parameter set is executed, as illustrated
in Fig. 5A. A lower amplitude EPSP will be produced
Summation of EPSPs and IPSPs can also be demonstrated using the model, as shown in Fig. 5B. In this
example, the transmitter released by cell A produces
an EPSP on cell C, whereas the transmitter released by
cell B produces an IPSP. If these events occur simul-
FIG. 6.
Effect of increasing membrane time constant on the shape (time course) of an EPSP triggered by cell A
(A). Dark line shows how both the time to maximum depolarization and the time to repolarization are
lengthened by increasing the membrane time constant of cell C from 0.4 (lighter trace) to 0.8 (darker
trace). B: the higher value for time constant results in temporal summation (dark trace) even at the
lowest firing rate. For comparison, the lack of summation during the identical pulse protocol is shown
by the lighter trace. Note: in both panels, the remember previous trace feature was turned on to overlay
the 2 traces.
VOLUME 25 : NUMBER 1 – ADVANCES IN PHYSIOLOGY EDUCATION – MARCH 2001
7
I N N O V A T I O N S
A N D
I D E A S
stant, the individual EPSPs do not summate, but they
progressively do as the time constant is increased
(arrow in Fig. 6B).
taneously (not shown), they will cancel exactly. If
they occur slightly out of phase (by increasing pulse
delay for cell B), this interaction becomes more obvious (Fig. 5B).
Presynaptic inhibition. At least two types of presynaptic inhibition can occur physiologically (3). In
both cases, a specific anatomical arrangement of
the presynaptic cells is required. This can be demonstrated (see Fig. 7) if the “wiring” control is
switched from “A-C” to “A-B,” meaning that cell A
now physically synapses on the terminal bouton of
cell B. In presynaptic inhibition, an action potential
traveling down the axon of cell A depolarizes the
terminal bouton of cell B (another “presynaptic”
cell with respect to cell C) just before an action
potential travels down the axon of B. The action
potential in cell A has no direct effect on cell C due
to the lack of a direct anatomical connection (not
shown here, but easily demonstrated using the
model). When the two presynaptic action potentials are timed appropriately, accommodation (a
form of voltage inactivation) may occur on cell B.
Effect of postsynaptic membrane time constant
on temporal summation. The value of the membrane time constant for cell C will determine the
shape and duration of the EPSP resulting from presynaptic stimulation. This time constant is determined by
a capacitance term, such that an increase in time
constant decreases both the rate of depolarization and
repolarization during the EPSP. Figure 6A illustrates
this effect by comparing the shape of an EPSP resulting from identical single pulses in cell A before and
after the time constant of cell C has been doubled
from its arbitrary initial value of 0.4 (arrow in Fig. 6A).
Postsynaptic neurons with high membrane time constants are known to exhibit enhanced temporal summation. This can be illustrated by using two pulses to
stimulate cell A. With lower values for the time con-
FIG. 7.
Demonstration of presynaptic inhibition. When “wiring” is changed to “A3 B,” the anatomical arrangement between the cells changes so that cell A synapses directly onto the terminal bouton of cell B. Thus,
when cell A fires alone, it does not produce an EPSP on cell C. When cell B fires alone, it produces a normal
2-mV EPSP on cell C. If cells A and B fire simultaneously (with A actually preceding B slightly), then the
amplitude of the EPSP produced by cell B is reduced (arrow). Note: the remember previous trace feature was
turned on to overlay the 2 traces.
VOLUME 25 : NUMBER 1 – ADVANCES IN PHYSIOLOGY EDUCATION – MARCH 2001
8
I N N O V A T I O N S
The result is that the amplitude of the action potential at the terminal bouton in cell B is diminished
slightly compared with normal, resulting in the
opening of fewer voltage-gated calcium channels
and less transmitter release from cell B. The resulting EPSP on cell C is thereby reduced in magnitude
relative to its normal value (arrow in Fig. 7). Presynaptic inhibition can result either from the opening of an excitatory receptor-gated channel on cell
B (as in Fig. 7) or from release of a transmitter such
as GABA that increases Cl⫺ conductance on the
terminal bouton of cell B and thereby anchors the
Em of cell B. These, again, are difficult concepts for
many students to appreciate from static diagrams,
and the model makes it easier to illustrate them.
A N D
I D E A S
cell C may reach the threshold for firing an action
potential. This is illustrated by the model as shown
in Fig. 8. With “# pulses” set to 20, both presynaptic cells firing, and the firing rate of one of the cells
set to a value of at least 9, an action potential is
triggered on cell C. This minimal combination of
temporal and spatial summation brings the Em of
cell C to threshold. The animation shows this by
briefly turning cell C dark at the point in time at
which the action potential is initiated (the flash
actually occurs at 2.6 ms under the conditions
shown in Fig. 8). The corresponding graph shows
the explosive change in Em for cell C during the
action potential. The entire action potential cannot
be shown on the same voltage scale required to
resolve 1- to 2-mV EPSPs and IPSPs, but its presence
is readily noted on both displays and is usually
effective at regaining students’ attention during a
lecture.
Events at threshold. When the controls of the
model are set appropriately so that sufficient summation of postsynaptic potentials occurs, the Em of
FIG. 8.
Behavior of model when summation of presynaptic input brings the Em of cell C to threshold for firing an
AP. The animation shows that cell C has fired an AP, and this is also reflected in the Em trace of cell C. The
lighter trace shows the result of settings that produced a subthreshold summation, using the remember
previous trace feature to overlay the 2 traces.
VOLUME 25 : NUMBER 1 – ADVANCES IN PHYSIOLOGY EDUCATION – MARCH 2001
9
I N N O V A T I O N S
A N D
I D E A S
DISCUSSION
IV. Start/Stop/Reset Directions:
This computer model simulates many basic physiological properties of the neuronal chemical synapse. A
number of possible uses for it are envisioned: as a
method to teach this subject to medical, undergraduate, or graduate students in a didactic setting, as an
alternative to a supervised animal laboratory, or as a
completely self-directed computer laboratory exercise.
Double-click on the “Synapse.rt” icon to load the program. After
it loads, the program runs once to redraw the two displays. Run the
program either by clicking on the “Run” button (3) in the upper
left-hand corner of the program window or by selecting “Run”
from the “Operate” menu at the top of the program page. The
program can be stopped at any time by clicking on the stop button
(stop-sign symbol) in the upper left-hand corner of the program
window. If necessary, the model parameters can be reset to their
initial conditions at any time by selecting “Reinitialize All to
Default” from the “Operate” menu at the top of the screen.
The model has been used in at least four courses at
Texas A&M University by faculty, medical students,
graduate students, and undergraduate students. In
each case, the student response has been positive,
and their understanding of this topic has increased.
To facilitate distribution and use of the model at other
educational institutions, it is compiled in Macintosh
and Windows versions and available for free download by faculty and students at the following internet
address: http://mphywww.tamu.edu/davis/models/
synapse.html. A comprehensive worksheet is provided in the APPENDIX and is also available for download as a Microsoft Word file.
Selecting “About” from the “Help” menu will pause the model and
bring up a series of informational screens about the model and the
experimental setup that would be used to make actual recordings
in live neurons.
V. Orientation and Brief Description of the
Model:
Load and start the program as described above. While the program
is running, but before beginning the laboratory exercise, look at the
computer screen and note that there are three general areas displayed on the screen. These areas include:
A. Animated Display of Neuronal Firing (top
left side of screen):
APPENDIX
On the left side of the screen is shown a pictorial diagram of the
anatomical connections between two presynaptic neurons (cells A
and B) and one postsynaptic neuron (cell C). A diagram of a
recording electrode placed near the axon hillock of cell C indicates
the site of membrane potential (Em) measurements.
Computer Laboratory Worksheet
(Synaptic Function)
I. Objective:
The default settings for the model are such that when the program
is first initiated, one presynaptic cell (cell A) will be stimulated with
one pulse at the lowest stimulation (firing) rate. Cells A and B each
synapse on the cell body of C. At any time, the user may pull down
the “Operate” menu at the top of the program page and select
“Reinitialize All to Default” to reset the program to the initial
conditions.
To help students better visualize basic principles of chemical synapse function.
II. Student Preparation for this Laboratory
Exercise:
It is assumed that students have attended lectures on the appropriate subjects or done preliminary readings before doing this exercise. Students should at least know the definitions of the terms
presynaptic, postsynaptic, temporal summation, and spatial
summation before doing the laboratory exercise.
B. Controls (lower left side of screen):
Below the animation are located various controls to alter the stimulus parameters, the anatomical arrangement between the cells,
and the membrane properties of the postsynaptic cell. Each presynaptic stimulus will produce an action potential in the respective
presynaptic cell that will travel down to that cell’s terminal bouton
and release transmitter.
III. General Information:
The simulation program name is “Synapse.rt” and is located on
PowerMac and Pentium-based computers in the College of Medicine, Learning Resource Center (LRC). The program may also be
downloaded from the World Wide Web for use at home. The
program may be run on computers with a processor speed of 40
MHz or greater.
The controls are organized such that there are four for each presynaptic cell, two for the postsynaptic cell (C) and two general
controls. The “Status” control for each presynaptic cell refers to
whether that cell is being stimulated (by an external stimulator that
is not shown) to fire an action potential. “Firing Rate” (rate)
VOLUME 25 : NUMBER 1 – ADVANCES IN PHYSIOLOGY EDUCATION – MARCH 2001
10
I N N O V A T I O N S
determines the interpulse interval if multiple stimuli are delivered
to the presynaptic cell (1 ⫽ lowest rate, 10 ⫽ highest rate). “Pulse
Delay” (delay) controls the timing between the stimulation of the
two presynaptic cells. “Time Constant” and “Length Constant”
control these properties for the membrane of cell C. “# Pulses”
determines the number of stimuli delivered to the presynaptic
cell(s) and can be used to fire a “train” of action potentials. “Wiring” controls the anatomical arrangement between the cells, i.e.
whether cell A synapses on cell C or on cell B. It is only used to
demonstrate presynaptic inhibition.
A N D
I D E A S
tion). Now change the time constant of cell C from 0.4 to 0.5. What
is the effect?
b. To observe the effect of changing time constant on a single
EPSP, RESET all variables, set A ⫽ fire, # Pulses ⫽ 1. RUN. Change
time constant and repeat. Describe how changing time constant
alters the EPSP magnitude:
and EPSP shape:
.
3. Spatial summation. This exercise demonstrates how stimulation of multiple presynaptic inputs can generate EPSPs that sum
over distance.
C. Em vs. Time Graph for Cell C (right side of
screen)
RESET. Compare the amplitude of the EPSP on cell C when A fires
one pulse vs. when A and B both fire one pulse in synchrony. A
alone (set A ⫽ fire):
A ⫹ B (set A and B ⫽ fire):
The display on the right shows the change in Em of cell C as a
function of time. This display provides quantitative information
about the impact of presynaptic stimulation on the postsynaptic
cell C. For this reason, it is more important than the animation
display, but many students are less confused if they can see that the
animation is synchronized with the change in the Em. The timing of
the presynaptic stimulus pulses is marked on the Em-vs.-time graph
by gray pulses just below the Em trace.
To more clearly observe that summation occurs, try increasing the
delay of cell B to ⬇ 1.4 ms: amplitude ⫽
.
At what value of Bdelay does no summation occur?
What is the maximum EPSP amplitude that can be achieved in this
model with spatial summation alone (i.e. without increasing the #
of pulses)?
Why?
VI. Laboratory Exercise:
4. Effect of Changing the Length Constant on Spatial Summation. RESET; A ⫽ fire, # Pulses ⫽ 1. Compare the EPSP amplitude with the length constant set to 1.0 vs. 0.1:
.
Set Pulses ⫽ 3 and Arate ⫽ 9, RUN. Compare the maximum EPSP
amplitude with length constant set to 1.0 vs. 0.1:
1. Temporal summation. This exercise demonstrates how stimulation of a single presynaptic cell generates excitatory postsynaptic potentials (EPSPs) that summate over time.
RESET all variables and set cell A ⫽ fire. Click the “Run” button.
Note the shape of the EPSP (a rapid rise in Em with exponential
decay back to the resting potential). What accounts for this shape?
[note: changing the length constant of cell C would not affect EPSP
amplitude directly under the terminal bouton of A, but remember
that the Em tracings are being measured at the axon hillock; therefore, changing the length constant affects how the amplitude of the
EPSP changes as it spreads from the bouton to the axon hillock].
Why isn’t the Em change a square-shaped pulse like the stimulus
pulse?
Increase “# Pulses” ⫽ 4 and run the model. Note that a train of four
action potentials (APs) travelling down cell A produces four individual EPSPs on cell C without any summation. Now sequentially
increase the firing rate of A from 1 to 9 and RUN the model after
each increase. Temporal summation should be observed; what is
the maximum amplitude of the summed EPSP?
At the highest firing rate (⫽9), how many pulses are required to
reach the threshold (⫺60 mV) for firing an AP on cell C?
(progressively increase the # pulses and RUN each time to find out;
alternatively, run the model in “Continuous Run” mode while
changing the number of pulses)
What should happen to the Em tracing if the threshold is
reached?
5. Combinations of Spatial and Temporal Summation. RESET,
set A ⫽ fire, Pulses ⫽ 5, Arate ⫽ 8, RUN.
What is the maximum EPSP amplitude?
What type of summation is involved?
Now add excitatory input from cell B (B ⫽ fire, Brate ⫽ 5). What is
the maximum EPSP amplitude now?
.
Progressively increase Brate to determine at what minimal rate an
AP fires on cell C:
.
Keep these settings and then increase Bdelay to 1.2 ms (this should
change the degree of summation slightly but enough to prevent an
AP from firing). What happens?
2. Effect of Changing the Time Constant on Temporal Summation. How does the time constant of the postsynaptic cell
influence the degree to which EPSPs summate on the surface of the
postsynaptic cell?
For another example of the interaction between temporal and
spatial summation, RESET, set A ⫽ fire, B ⫽ silent, Arate ⫽ 9,
Pulses ⫽ 5, tau ⫽ 0.6 then RUN. Note that threshold is not quite
reached, but if cell B fires even once (set B ⫽ fire to see this), an AP
fires on cell C.
a. Keep all variables the same as they were at the end of exercise
1 (i.e. a rapid train of pulses firing on cell A that gives a large but
NOT-QUITE-threshold EPSP on cell C (a 10- to 12-mV depolariza-
[This shows how cell B could act as an ON-OFF gate to control
“information” transfer from cell A to cell C, i.e. if B is off, then
repeated bursts of firing by cell A do not elicit APs from cell C, but
VOLUME 25 : NUMBER 1 – ADVANCES IN PHYSIOLOGY EDUCATION – MARCH 2001
11
I N N O V A T I O N S
A N D
I D E A S
Now set A ⫽ silent, B ⫽ fire. You should see a “normal” EPSP on
cell C. What happens when cell A fires also (set A ⫽ fire)?
What mechanism underlies this phenomenon?
Keep the settings the same. What happens if the Bdelay is progressively increased to 3.0?
Why?
if B fires a pulse during that time (Bdelay ⫽ 1.0 –2.0 ms), an AP fires
on C. Try to verify this].
6. Effects of Changing Postsynaptic Conductance (transmitter/receptor type). [it may be useful to run the program in
continuous loop mode for exercises 6 and 7].
a. RESET, set A ⫽ fire. Measure the amplitude of the EPSP:
and the default conductance (cond) setting:
The author thanks Dan Phillips at National Instruments for easing
the licensing restrictions and thereby allowing free distribution of
reasonable numbers of this program, along with complied LabView
libraries, to faculty and students who contact this website. Judy
Davidson provided assistance with proofreading and printing.
Medical and graduate students at Texas A&M College of Medicine provided valuable assistance in testing early versions of the
model.
This could simulate the effects of ACh as the transmitter released
from cell A, because ACh typically increases the membrane conductance of the postsynaptic cell to both K and Na ions. To
illustrate how this can cause a depolarization (despite K efflux),
compare the size of this EPSP with one resulting from a selective
increase in gNa (i.e. change Acond ⫽ “Na”):
b. Now set Acond ⫽ “Cl” and note the magnitude of the
EPSP:
Explain the difference:
Compare this with the inhibitory postsynaptic potential (IPSP)
amplitude when Acond ⫽ “K ⫹ Cl”:
or when Acond ⫽
“K”
If changes in gCl have “no effect” on Em, why is the IPSP amplitude
smaller when Acond ⫽ “K ⫹ Cl” than when Acond ⫽ “K” alone?
Address for reprint requests and other correspondence: M. J. Davis,
Rm. 346 Reynolds Medical Building, Texas A&M University HSC,
College Station, TX 77843.
7. Presynaptic inhibition. Demonstration of this phenomenon
requires an anatomical change in the configuration of the cells.
RESET all variables, then set “Wiring” ⫽ A 3 B and RUN once.
1. Hogdkin AL and Huxley AF. A quantitative description of
membrane current and its application to conduction and excitation in nerve. J Physiol (Lond) 117: 500 –544, 1952.
2. Katz B. Nerve, Muscle and Synapse. New York: McGraw-Hill,
1966.
3. Matthews GG. Cellular Physiology of Nerve and Muscle. Palo
Alto, CA: Blackwell, 1986.
Received 23 February 2000; accepted in final form 7 December
2000
REFERENCES
a. Set A ⫽ fire, B ⫽ silent. Now what effect does cell A firing have
on the Em of cell C?
Why?
VOLUME 25 : NUMBER 1 – ADVANCES IN PHYSIOLOGY EDUCATION – MARCH 2001
12