Chapter 3: The synapse – regulating communication and modifying
In the previous chapter we discussed the neuron – the fundamental unit of the nervous system.
We learned about its key role in information flow:
It transmits an electrical signal from the input site, to the site where the signal will
be conveyed onward to its partner cell.
We learned the strategies the axon uses to maximize the efficiency with which the electrical
signal is conveyed.
We examined the transport mechanisms in place to support information flow. The structural
proteins the neuron requires are synthesized in the cell body. Hence ensuring that the
dendrites, axon and presynaptic terminal are supplied with the right material at the right time
requires a complex transport system. We learned how microtubules and ATPase motors
collaborate to transfer cargo where and when it is needed.
Finally we learned what may go wrong if either the electrical signaling or the support systems
are abnormal. We learned how information flow can be disrupted:
Disorders of electrical signaling such as those that affect myelin, can directly affect
the speed that the electrical signal is conveyed.
Disorders of transport mechanisms, such as those that affect the nerve terminal,
can directly affect how information is conveyed at the synapse.
Disorders of transport mechanisms, such as those that affect axon structure, can
indirectly affect the speed that the electrical signal is conveyed, because myelin
only remains intact when the axon has a healthy diameter.
In this chapter we are going to focus on the region where the electrical signal is transformed into
a chemical signal to travel across the gap between a neuron and its partner – and then we will
examine two important disorders in which signaling across this gap – the synapse – breaks
The synapse - where electrical information is converted to a chemical signal.
We will begin by describing one of the best-characterized synapses – the vertebrate
neuromuscular junction – where spinal motor neurons communicate with skeletal muscle.
Once we understand this synapse, we will be better prepared to understand how other neurons
have modified this basic structure to meet their specialized requirements.
The vertebrate neuromuscular junction: The neurons that innervate skeletal muscle are
called motor neurons and their cell bodies are found in the motor (ventral) region of the spinal
cord. They send their axons in bundles to contact muscle cells in the periphery. Because the
motor neuron axons are long, they need to be heavily myelinated. As a general rule, each
muscle fiber receives input (is innervated) from a single motor neuron, but each neuron in the
bundle can synapse onto more than one individual muscle fiber. Together the motor neuron and
its target muscle fibers create a motor unit that can vary in size from just a few fibers in the
small muscles used for fine movements, to up to ~100 fibers in the large muscles used for gross
The neuromuscular synapse is excitatory,
that is to say, transmitter release from the
presynaptic terminal causes the membrane
of the muscle fiber to depolarize, eventually
causing the muscle to contract.
acetylcholine (ACh), as the chemical
neurotransmitter to communicate with the
muscle. In the presynaptic terminal Ach is
found in small organelles called synaptic
vesicles. The synaptic vesicles cluster
around specialized sites in the presynaptic
plasma membrane called active zones,
where the vesicles can fuse with the
presynaptic plasma membrane. When they
do fuse neurotransmitter is released into
the synaptic cleft – the extracellular space
between the presynaptic site at the axon
terminal and the postsynaptic site on the
muscle. Directly opposite the active zone,
the postsynaptic muscle cell membrane
holds the receptors that bind Ach. They are
clustered at a very high density (about
10,000 to 20,000 receptors/μm2, which is
almost a crystalline array). The synaptic
cleft contains a high concentration of
acetylcholinesterase, the enzyme that
hydrolyzes and inactivates the Ach after it
has been released. The cholinesterase is
held in place by a collagen matrix on the
surface of the muscle.
Figure 1: The vertebrate neuromuscular junction.
1A: The motor neurons arise in the motor region of the spinal cord and project out to the skeletal muscles.
1B: Each axon makes many terminal branches, each of which ends in a synapse. The motor neuron uses
acetylcholine (Ach) as a neurotransmitter.
IC: Ach is synthesized in the terminals and loaded up into synaptic vesicles. When the action potential reaches
the terminal it causes vesicles to release Ach into the synaptic cleft. Ach that binds to its receptor on the
postsynaptic site depolarizes the postsynaptic membrane, causing a synaptic potential. At healthy synapses,
every synaptic potential gives rise to a muscle action potential.
The presynaptic side: How neurotransmitter is released.
When the action potential moving down the axon of the
motor neuron invades the presynaptic terminal, it
activates voltage-gated Ca2+ channels localized at the
active zones in the terminal membrane. They open, and
Ca2+ floods into the terminal, triggering the fusion of
synaptic vesicles with the plasma membrane and the
release of transmitter into the synaptic cleft. Early
investigators, recording electrical signals from the
synaptic contacts on muscle, observed that Ach could be
released even in the absence of an action potential. But if
the nerve is not stimulated, the probability of a vesicle
fusing with the presynaptic terminal and releasing
transmitter is very low. This is because the release
apparatus that controls fusion at the active zone is
composed of proteins that depend on Ca2+ to become
activated, and in the absence of an action potential the
free Ca2+ concentration at the active zones is normally
very low. When the action potential enters the presynaptic
terminal and causes voltage-dependent Ca2+ channels to
open, it increases Ca2+ levels by 1000 fold.
Whenever a single vesicle fuses with the plasma
membrane and delivers its contents into the synaptic cleft,
it releases the same amount of transmitter into the cleft.
The amount released is called a quantum. Each quantum
contains approximately 3,000 molecules of ACh. A
quantum is the minimum amount of Ach able to elicit a
single synaptic potential on the post-synaptic side. Many
synaptic potentials have to occur to cause a muscle
action potential, but in healthy neurons each action
potential arriving at the terminal stimulates enough
transmitter release to cause a muscle action potential.
Figure 2: The Presynaptic side of the neuromuscular junction: a) A diagram of the active zone. The (purple)
synaptic vesicles line up next to an area where proteins of the fusion/releasing complex are concentrated
(yellow). Once the Ca2+ influx has activated them the vesicles dock and releases Ach into the cleft through a
pore (green). b) An actual electron micrograph of a neuromuscular synapse. The synaptic vesicles are arrowed.
Notice the fibrous material inside the synaptic cleft where the acetylcholinesterase enzyme is located. c) An
electron micrograph that has caught some of the synaptic vesicles in the process of fusion and releasing
transmitter into the synaptic cleft. d) Looking up at the active zone of a presynaptic terminal from the
presynaptic cleft using a type of electron microscopy that is sensitive to the membrane. The ‘beads’ on the
surface are membrane proteins while the ‘dimples’ show pores in the membrane where the synaptic vesicles
have fused. After releasing transmitter the pore closes and the vesicles bud off inside the terminal where they
can be refilled and reused.
The postsynaptic side: How the chemical signal is interpreted.
Once ACh is released into the synaptic cleft, two things can happen: It will either bind to
nicotinic receptors on the muscle membrane, or it will be inactivated by the cholinesterase in
the cleft. The activity of the cholesterase enzyme is one of the highest known for any enzyme;
about half of the ACh is broken down even before it reaches its receptor. ACh that does reach
the receptors binds (two molecules per receptor) and triggers a conformational change so that
the receptor opens a water-filled pore, or channel, that spans the muscle membrane.
This receptor channel is different from the voltage-gated channels involved in the action
potential because it is permeable to both Na+ and K+. Thus, when the channel opens:
Na+ flows into the cell down its electrochemical gradient.
K+ flows out of the cell down its gradient.
There is more flow of Na+ than K+ ions through the channel, resulting in a potential change
across the muscle membrane, called the synaptic potential. It causes the muscle membrane
The more Ach that is released by the
presynaptic terminal, the more Ach
receptor channels are activated, the
larger the synaptic current, and the
larger the synaptic potential. But unlike
action potentials, synaptic potentials
are not regenerative. Their amplitude
decreases with the distance from the
region of synaptic contact. Thus,
synaptic potentials are confined to the
membrane at or near the synapse.
Figure 3: The acetylcholine (Ach) receptor. a, b) Models of
the acetylcholine receptor complex a) from the top, b) from the
side. It is a pentamer whose subunits surround a central pore,
like a donut. When two molecules of Ach bind to the receptor,
the ensuing conformational change causes the pore to open.
Na+ ions move into the muscle cell and K+ ions move out. c) A
model of the receptor from the side, cut away to show the pore.
The ‘gate’ is made from negatively charged amino acids that
surround the pore and ensure that it is only permeable to
cations. The location of the receptor in the membrane is
For muscle contraction to occur the
synaptic potential has to be translated
into an action potential. This happens
when enough ACh is released from
the nerve terminal to depolarize the
muscle cell past the threshold that will
activate the voltage-dependent Na+
channels in the muscle membrane.
When these voltage-gated Na+
channels are activated they generate
a muscle action potential that is
propagated from end to end in the
muscle fiber, causing contraction. For
this reason, the synaptic potential is
It is important to note that the AChactivated Ach receptor channels are
not themselves directly involved in
generating the action potential; they
just provide the stimulus current that
depolarizes the membrane enough to activate the voltage-dependent Na+ channels in the
At normal, healthy nerve-muscle junctions, neurons always release enough ACh to produce a
synaptic potential that is above the threshold that will give rise to a muscle action potential.
Thus, every time the nerve is stimulated, the synaptic potential will cause a twitch contraction
response in the muscle. Neuropathies in which Ach release is reduced result in decreased
synaptic potentials so the muscle membrane doesn’t always reach threshold. In this case,
activation of the neuron may not lead to muscle contraction, and synaptic transmission fails.
Other Excitatory Synapses
Nicotinic Ach receptors produce excitatory synaptic potentials because they cause the
postsynaptic membrane to depolarize. This is true whether the membrane the receptors are in
skeletal muscle membrane, or another neuron. Nicotinic Ach receptors are also found on
autonomic and central nervous system neurons so they cause excitatory synaptic potentials
The most widespread receptors involved in excitatory
synaptic transmission in the central nervous system are
the Glutamate receptors. They exist in many
functionally distinct forms and play a critical role in
learning and memory. Both Ach and glutamate are
known as classical excitatory neurotransmitters
because they are small organic molecules.
peptides can also act as excitatory transmitters.
For example, the peptide substance P is released from
sensory nerve terminals and is a key signal in pain
pathways. As with the classical neurotransmitters, each
peptide neurotransmitter activates its own unique
receptor and gives rise to a characteristic synaptic
Figure 4: Glutamate receptors. Glutamate synapses are the most common excitatory synapses in the central
nervous system. In this picture of a neuron from the hippocampus each fluorescent blue spot marks a glutamate
synapse. Note how abundant they are on both the dendritic tree and the neuronal cell body.
Central Synapses are Different in Some Ways from the Nerve-Muscle Junction
The neuromuscular junction evolved to perform the task of
providing a rapid and high fidelity response to activation of
motor neurons. Because of this it has many structural
features - such as a presynaptic terminal with a high
probability of transmitter release and a highly sensitive
postsynaptic membrane, specifically designed to carry out
such a charge. Synapses in the central nervous system,
however, are designed for different functions - not as
slaves to a single presynaptic input, but as integrators of
information from many inputs that may be both excitatory
and inhibitory. As you might anticipate, the structural
features of such synapses differ in important ways from
the nerve-muscle synapse. Their structure is much more
rudimentary, with fewer presynaptic vesicles, less
organized active zones, and simpler postsynaptic
specializations. In addition, a
single postsynaptic cell receives Figure 5: Central Nervous System (CNS) synapse. Many CNS
many (hundreds or even synapses are located on protrusions from the dendrite shaft. These
protrusions increase the surface area of the neuron. Many CNS
neurons have thousands of synapses on their dendrites and cell
thousands) of presynaptic inputs, and some of these can be switched on and off, as
Figure 6: Central Nervous system (CNS)
Inhibitory as well as excitatory synapses.
This is a pseudocolored electron micrograph
of two types of CNS synapse. Two
presynaptic terminals are shown. The one in
pink synapses on a dendritic spine, whereas
the one in blue synapses on the dendritic
shaft itself. Usually excitatory synapses are
on the spines, while inhibitory synapses are
on the shaft. The synaptic vesicles are shown
in purple. The actual synapse has been
colored green – note that the pink terminal
makes two excitatory synapses on the
dendritic shaft. This will allow summation of
the signal. Unlike at the neuromuscular
junction the postsynaptic side of the CNS
synapse is less specialized. Dark electron
dense material indicates where the receptors
are located. Excitatory synapses have more
electron dense material than inhibitory
The neuromuscular junction and other excitatory synapses depolarize the postsynaptic
membrane above the threshold required to elicit a postsynaptic action potential. In contrast an
inhibitory synapse suppresses the action potential. As you might imagine, inhibitory
neurotransmission is somewhat different than excitatory.
Many inhibitory neurotransmitters are known. In the central nervous system the most important
is Gamma-aminobutyric acid, more commonly referred to as GABA. GABA regulates many
functions including controlling movement by regulating input to the motor neurons in the spinal
cord. GABA has many different types of receptors in the nervous system, but the most important
is the GABA(A) receptor. Structurally it resembles the nicotinic ACh receptor – when two GABA
molecules bind they trigger a conformational change and the receptor opens a water-filled pore,
or channel that spans the postsynaptic neuronal membrane. In contrast with the nicotinic Ach
receptor, this channel is permeable only to chloride (Cl-) ions. This has several
consequences on the response of the postsynaptic membrane:
Since Cl– is concentrated outside of cells, opening of GABA-gated channels results in
an influx of Cl– into the postsynaptic cell.
This causes the postsynaptic membrane to hyperpolarize.
As a result of this hyperpolarization, the postsynaptic membrane is driven away from the
threshold required to activate voltage-gated sodium channels.
The postsynaptic cell is therefore inhibited from firing action potentials.
Drugs that target GABA(A) receptors, such as
benzodiazepines (e.g., valium) work by increasing the activity
of GABAA receptors. This has the result of suppressing
excitatory nervous system activity in the brain. The concept
that the GABA receptor needs to be more active to produce
more inhibition is an important one.
Other inhibitory synapses are also important. The glycine
receptor is one example. Like the GABA(A) receptor it has a
Cl- channel. Glycine has an important role in integrating
control of movement. Both GABA and glycine receptors have
been implicated in spastic disorders.
Slow as well as fast effects
and Figure 7: The GABAA receptor. Like the Ach receptor the GABAA
inhibitory receptor types we receptor is a pentameric protein with a pore down the center, like a
discussed donut. Like the Ach receptor, 2 molecules of GABA bind and
above are protein complexes trigger a confromational change that opens the pore. Unlike the
with integral ion channels – the Ach receptor the GABAA channel is permeable to an anion, Cl-,
only differences are the types of which flows into the postsynaptic cell and causes the membrane to
hyperpolarize, therefore inhibiting generation of an action
ion conveyed through the pore. potential. Many drugs that inhibit nervous system activity bind to
As a consequence these receptors the
extremely quickly when the neurotransmitter binds.
But in other cases the part of the receptor that binds the transmitter is not physically associated
with the ion channel. In these cases, when a transmitter binds to its receptor an intervening
‘second messenger module’ conveys information to the channel. These second messengers
are familiar from their other functions in cells and include cyclic AMP, cyclic GMP, Ca2+,
calmodulin, and GTP-binding proteins. They have the flexibility to exert a number of different
effects. For example, they may:
Bind directly to a channel, gating it open or closed, or changing its opening probability.
Activate another effector molecule (such as a kinase or phosphatase enzyme) that, in
turn, alters channel function.
Have effects not directly related to channels – such as on gene transcription.
Figure 8: A slowly responding receptor. A
diagram showing the typical arrangement of a
receptor in which the ion channel is not physically
associated with the receptor binding protein. The
transmitter binding module is usually a 7
transmembrane protein that is coupled to a G
protein. The G protein in turn activates a second
messenger module. Depending on whether the G
protein is inhibitory or stimulatory, and the type of
2nd messenger, the channel may be opened or closed.
Then, depending on the type of channel, the
membrane may be depolarized or hyperpolarized.
by a neurotransmitter
will respond much faster than an
ion channel that is indirectly activated via a second messenger module. Many transmitters in the
CNS have both fast and slow receptors. Because of this the outcome of a synaptic event
depends more on which receptor is present than on the transmitter itself. For example,
Ach has both fast (nicotinic) and slow (muscarinic) receptors. The muscarinic receptors can
have excitatory or inhibitory effects, depending on the second messenger module they are
coupled to. Therefore, Ach can have either fast excitatory, slow excitatory, or slow inhibitory
effects, depending on location.
The combination of excitatory and inhibitory synapses and the differences in speed of response
means that the central nervous system has a wide range of possible ways to modulate neuronal
behavior, as we shall see below.
Remember! a single neuron in the central nervous system can receive thousands of inputs,
which are often conflicting and overlapping. How does each CNS neuron integrate many inputs
into a coherent output? Synaptic connections occur on the dendrites and cell bodies of neurons.
The cell body of the neuron integrates all of the synaptic inputs. The goal of input integration is
to put the postsynaptic neuron into a final electrical state whereby it can either fire an action
potential or not.
How synaptic input onto a neuron is integrated.
One way is by inhibiting output from a synapse. We
have discussed how inhibitory synapses function on
the postsynaptic side of the synapse (postsynaptic
inhibition). Another critical way that inhibitory
synapses influence output is to interfere with
transmitter release from the presynaptic nerve
Presynaptic inhibition uses a third modulatory
neuron to regulate the behavior of a normally
excitatory synapse. In this case a presynaptic
inhibitory neuron releases a transmitter onto the
presynaptic nerve terminal. The transmitter binds a
receptor in the presynaptic terminal that uses a second messenger module to inhibit the
voltage-dependent Ca2+ channels that are required to cause vesicle fusion. As a result of this
inhibition the amount of Ca2+ entering the presynaptic terminal during the action potential is
reduced, and this in turn reduces the amount of transmitter released, preventing the
postsynaptic membrane the reaching the threshold that will cause a synaptic potential.
The axon will fire an action potential if:
The excitatory inputs are greater than the inhibitory inputs.
It will not fire an action potential if:
The inhibitory inputs are greater than the excitatory inputs.
Integration of synaptic potentials can be both spatial and temporal
Synaptic potentials decay with distance from the synapse. Thus, an input far away from the cell
body will be less effective in controlling the fate of the neuron than an input close to the zone
where the action potential initiates. Hence, both the amplitude of the synaptic potential and its
point of origin relative to the action potential initiating zone are important determinants of
whether it will activate the axon.
Synaptic potentials are relatively slow events, lasting tens of milliseconds. So if the same
excitatory presynaptic input is stimulated more than once within a short period of time it
will produce postsynaptic potentials at the same point on the postsynaptic cell that can
sum with one another to produce a larger synaptic response than either one alone would
produce. Figure 6 has an example of a synapse that could behave in this way. This
'summation in time' is termed temporal summation.
If two independent presynaptic inputs are stimulated at exactly the same moment in
time, to produce two synchronous postsynaptic potentials at different points on the
postsynaptic membrane these potentials will sum to produce an integrated synaptic
potential as they spread passively toward the action potential initiation zone at the start
of the axon. This summation in space is termed spatial summation.
If both synaptic potentials are the same type (both excitatory or both inhibitory) then the
summed potential will be larger than that produced by either one alone. On the other
hand, if one synaptic potential is excitatory (depolarizing) and the other (from a different
presynaptic terminal) is inhibitory (hyperpolarizing), then summation of the two will lead
to a synaptic potential whose amplitude is intermediate between the two.
These simple principles of synaptic integration are in continuous operation in every neuron in
the nervous system. Each cell integrates all of the synaptic information impinging on it and,
depending upon the balance of excitation and inhibition, it either fires an action potential or it
does not. These are the basic means by which all observable behavior is controlled in an
Figure 10: A. Spatial and temporal
summation. B. Temporal summation occurs
when the same excitatory input fires close
together (like the synapse in Fig 6). C.
Spatial summation occurs when different
excitatory input fires close together. D.
Excitatory and inhibitory input can cancel
each other out.
We will now turn our attention to two types of disorders that exemplify the synaptic
behaviors we have discussed above.
Pain is clearly a sensory experience with a crucial protective function; it warns of injury and in its
absence the body cannot avoid damage. But there is a key distinction between the neural
mechanisms by which we sense pain – called nociception - during which specialized sensory
receptors called nociceptors are activated by noxious insults, and pain itself - which is the
response to actual (or even perceived) tissue damage. The distinction is clinically and
experimentally very important. Nociception does not necessarily lead to the perception of pain.
In fact the intensity with which pain is felt depends on the individual and the surrounding
conditions almost as much as the sensory stimulus itself. Hence there are no ‘painful’ stimuli
that invariably elicit the perception of pain in all individuals. The highly subjective nature of pain
is one of the factors that make it difficult to define and treat clinically.
There are different categories of pain
In order to understand how we perceive and respond to noxious stimuli we must be aware of the
distinct categories of pain:
1. The major reason patients usually seek medical attention is because of persistent pain.
This is pain with an identifiable cause. It can be divided into two categories:
Nociceptive pain is usually caused by inflammation and results from direct activation of
sensory nociceptors in the skin or soft tissues.
Neuropathic pain is caused by direct injury to nerves and bypasses the nociceptors.
2. The reason patients seek medical attention is because of chronic pain, which is very real
even though there is no obvious underlying cause. It can be a significant clinical burden.
Pain pathways deliver nociceptive information from the periphery to the brain where it either
will or will not be perceived as pain. The ascending pathways headed toward the brain have
three important synapses, each of which plays a different role in transmitting the signal to the
1. At the periphery where the nociceptive response is activated.
2. In the spinal cord where nociceptive information is gathered into distinct pathways that
ascend towards the brain.
3. In the cortex where nociceptive information is integrated with input from emotional
centers, and potential responses are assessed.
The brain’s response to nociceptive information has two components:
Non-specific motor output that initiates avoidance.
Pain-specific outputs that initiate analgesia.
We will now take a look at how each of these important synapses function in the pain response.
1. The first pain synapse in the periphery
Nociceptive information from the skin and soft tissues: While mechanosensory information
from the skin is recognized by a number of specialized receptors, such as Pacinian corpuscles
that are responsive to pressure and Merkel’s discs that are responsive to vibration, noxious
information is recognized by nociceptors that are simply free nerve endings located in the
Figure 11: Nociceptors in the skin. In
contrast with mechanoreceptors that are
associated with myelinated Abeta (A
fibers, nociceptors are simply free nerve
endings. However they can be associated
with either myelinated Adelta (Aor
non-myelinated C axons.
We can identify three types of skin
nociceptor based on their distinct
Thermal nociceptors are activated by extreme temperatures (>45°C or <5°C). They
transmit information quickly because their Adelta (A axons are myelinated.
Mechanical nociceptors are activated by intensive pressure. They also transmit
information quickly via myelinated A axons.
Polymodal nociceptors are activated by high-intensity mechanical, chemical or thermal
stimuli. They transmit information more slowly because their C axons are nonmyelinated.
Nociceptive information from the internal organs is detected by receptors that are activated
by inflammation and chemical insults. They are called ‘silent nociceptors’. We shall see why in a
How are nociceptors activated?
Nociceptors in the skin are chiefly
activated by tissue damage that results
in inflammation. The inflamed area
bradykinin, which stimulates the
nociceptors to release substance P.
The first pain synapse in the periphery. The
free nerve endings are stimulated when the
skin is inflamed. This sets up a reflex that
amplifies the noxious signal. i.e. the axon does
not need to be depolarized for the response to
occur. Once the signal is over threshold it is
conveyed along the nerve to the next synapse,
in the spinal cord.
neurotransmitter with an excitatory receptor. The substance P released by the nerve terminal
reacts with receptors on mast cells, causing them to degranulate and release histamine. The
histamine can excite the nociceptors too, amplifying their response. Histamine also causes
blood vessels to dilate and plasma to leak out of capillaries causing swelling (edema). This is
known as the triple response of Lewis. Once the axon reflex has amplified the signal above
threshold the nociceptive neurons send the noxious information back to the spinal cord. Note
that noxious means any painful stuimulus not simply chemical.
Nociceptors will transmit noxious information
quickly (if myelinated A fibers are activated) or
slowly (if non-myleinated C fibers are activated).
This impacts how pain is felt: At first, a sharp
sensation is due to the A fiber response, then a
second burning or aching pain due to the C fibers.
Fast pain is due to activation of myelinated Ad
Slow pain is due to activation of unmyelinated C
2. The second pain synapse in the spinal
The role of the spinal cord is to convey
information to and from the brain. It is divided
into two general regions:
Gray matter – where the synapses
White matter – where the ascending
and descending pathways are.
The gray matter is divided into two main
The sensory area is at the back
The motor area is at the chest
Incoming nociceptive fibers synapse in the gray
matter of the sensory area (dorsal horn)
The gray matter is further divided into a number of layers (laminae). The pain synapses are
located in the sensory area, in layers I–VI. Where the myelinated A fibers and non-myelinated
C fibers synapse turns out to be crucial for how the initial noxious sensation is controlled.
The fast A fibers synapse with projection (output) neurons that move the information
up the spinal cord to the brain
The slow C fibers synapse with both projection (output) neurons and with interneurons
that then synapse on projection (output) neurons.
Pressure sensitive Abeta (A fibers (mechanoreceptors) also synapse with the
projection (output) neurons.
This circuit in the spinal cord is the first way we manage the output of noxious information.
1. Normally the interneurons synapse on
projection neurons (that output to the brain)
inhibiting them from firing.
2. The C fibers synapse on both the inhibitory
interneurons and the projections neurons.
3. When the C fibers are activated by a noxious
interneurons. This allows the output neurons
to fire when they are activated by A fibers in
response to the painful stimulus.
4. The C fibers also stimulate the projection
neurons directly – so even more pain
information is transmitted up the spinal cord.
5. However, the inhibitory interneurons also
receive input from mechanosensitive A
fibers. Unlike the C fibers, the A fibers
stimulate the inhibitory interneurons.
6. This means that when the A fibers activate
the inhibitory interneurons the projection
neurons are inhibited - this prevents the
noxious stimulus being transmitted to the
Thus if we can tip the balance toward activating the
Ab fibers, we will feel less pain – this is why we rub
ourselves after we have experienced a painful
Mechanosensitive AB fibers regulate
nociceptive output at the spinal cord.
Chronic pain occurs when C fibers are persistently activated.
It is easy to see therefore why chronic activation of C fibers would cause a pain to persist even
after the initial stimulus from the periphery has been dealt with. The inhibitory interneurons can
no longer work. Persistent changes in C fibers behavior can cause perception of pain even
when the initial injury has healed.
Nociceptive axons from both the skin and the
internal organs project to the same output neurons
in the sensory region of the spinal cord. But
sensations from the skin normally predominate.
Hence, when the nociceptive axons from the internal
organs are activated, higher brain centers
incorrectly localize the sensations to different areas
of the skin. Thus injury to an internal organ is
Nociceptive neurons from the skin and the internal
organs synapse in the same place in the spinal cord
– the brain cannot tell where the stimulus is coming
experienced on predictable areas of the body surface - when you feel a pain in the arm following
a heart attack, this is because the brain is misinterpreting the source of the painful stimulus, not
because your arm has been damaged. Why are these areas so predictable? Because pain
neurons from the skin and the internal organs are always coupled in the same area.
The inability of the brain to directly recognize nociceptive stimuli from the internal organs is why
these receptors are called “silent”.
Referred pain has a stereotyped distribution
The stereotyped distribution of referred pain is used to diagnose damage to internal
3. The third pain synapse - How nociceptive information gets from the second
synapse to the cerebral cortex.
The pain synapses in the sensory regions of the spinal cord gather neurons together to send
nociceptive information to the brain using three important pathways.
The most prominent pathway is concerned with where the pain is localized. It ascends
from the spinal cord to the thalamus, which is the important way station for all neurons
destined for the cortex. From the thalamus the pain pathway ascends to the somatic
sensory cortex. Almost as soon as the pathway starts, the axons cross to the opposite
side of the spinal cord. Thus the information is represented on the opposite side of the
The second pathway is concerned with the non-specific arousal that occurs after noxious
stimulation occurs. It too goes to the cortex, but also branches into the area in the
brainstem that organizes arousal.
The third pathway ends up in an area of the brain stem that can initiate analgesia. We
will look at how this happens shortly.
How the cerebral cortex processes nociception
The primary nociceptive
pathway from the spinal
cord ends up in the
opposite part of the
cortex that recognizes
sensation – the parietal
cortex. The pink line on
the model of the cortical
lobes is the location of
the primary somatosensory cortex. If we
slice down this line and
lie the slice flat we can
see how the body is
mapped in this area of
the cortex. It is clear that
some areas of the
Sensory input from the body maps onto the parietal cortex at the ‘somatosensory
over- strip’. The homunculus reflects the differences in sensory input from each area.
represented on the
map, indicating that they send a lot more sensory information to the cortex. Nociceptive
information is processed in this area along with other sensory information. It is clear from this
map that pain does not simply arise from how nociceptive information is processed in the
somatosensory cortex. If it did, the sensation would reflect the small, well-defined, receptive
fields of the nociceptors. Instead most clinical pain involves diffuse aches. These reflect the
involvement of other brain areas. For example:
The insular cortex is found directly underneath the primary
somatosensory cortex. It processes information about the
internal state of the body and contributes to the emotional
response to pain. Patients with lesions in the insular cortex are
emotionally unresponsive to pain.
The cingulate gyrus is part of the limbic system that
is also important in emotional responses.
Errors in processing – phantom pain.
Phantom pain occurs almost exclusively as a result of amputation. Almost immediately following
the amputation of a limb, 90-98% of patients report experiencing a phantom sensation. Nearly
75% of individuals experience
the phantom as soon as
anesthesia wears off, and the
remaining 25% of patients
experience phantoms within a
few days or weeks. For some
disappear or change over
time, in others they may
continue for years or even a
lifetime. Phantom limbs are
very diverse and individual. The solid lines show the site of amputation, the dotted lines
Some describe their phantom where the phantom limbs were experienced. Nearly 95% of
limbs as being stuck in a fixed people report a phantom limb experience.
position whilst other claim to
be able to move their limbs both voluntarily and spontaneously to the extent that they even
‘gesticulate’ when they talk. Amputees often describe that parts of their limb are missing or that
the limb is magnified, stretched, or shortened. The posture the phantom ‘adopts’ is often related
to the last pose the individual saw their limb in before it was amputated, and this position is
often associated with pain.
Little is known about the true mechanism causing phantom pains. Historically, phantom pains
were thought to originate from nerve scars (neuromas) located at the stump tip, but this does
not explain why the sensations appear to emanate from within the space occupied by the
missing limb (rather than from the stump).
Errors in processing
As we saw previously, when a limb is stimulated the
corresponding part of the opposite somatosensory cortex
is activated. When a limb is removed this part of the brain
no longer receives its normal activation. This causes the
brain to reorganize so that the part of the brain that used
to respond to the missing limb responds to other things
instead. For example, it may respond to touch to a
different part of the body such as the face. The standard
explanation of phantom limb sensation is that the part of
the brain that used to respond to the limb is now
responding to other things: such as different parts of the
body, or maybe to the sight of that part of the body.
The pink line shows the area of
somatosensory cortex no longer
receiving input if the right arm is
Why are some phantom limbs painful?
Some people have painless phantom limbs, whereas
others experience excruciating pain. We don’t
understand why. One factor known to be important is
whether the limb was painful prior to amputation. If the real limb was painful prior to amputation
then there is a higher chance that the phantom limb will be painful too. One suggestion is that
the more that the brain reorganizes itself after amputation, the more pain will be experienced.
For most people the intensity of pain and the actual nature of the phantom limb can also change
over time. One common phenomenon is called 'telescoping' in which the phantom arm or leg
appears to get shorter over time.
Mirror box therapy.
Many patients experience pain because the phantom limb seems to be clenched; obviously
phantom limbs are not under voluntary control, so unclenching is impossible. The theory behind
mirror box treatment is that the brain has become accustomed to the fact that a phantom limb is
paralyzed because there is no feedback from the phantom back to the brain to inform it
otherwise. The neurologists Ramachandran and Rogers-Ramachandran believed that if the
brain received visual feedback that the limb had moved, then the phantom limb would become
To create the visual feedback that
would allow this illusion, they
constructed mirror boxes that
have a vertical mirror placed in
the center. The intact limb is
placed on one side of the mirror,
in the patient’s sight, while the
amputated limb is placed on the
other side, out of sight. The
patient sees the intact second
limb through the mirror and sends
commands to both limbs to make
movement gives the brain
phantom has moved, and it
In a study of ten patients with
upper phantom limb paralysis, nine patients were able to ‘move’ the phantom limb, and eight of
those patients had their pain alleviated. Since this pioneering study, multiple additional studies
have support the mirror box findings for patients with upper limb phantom pain. The first case of
mirror box treatment for lower limb phantoms was reported in 2004. The patient, Alan,
experienced a painful crossing of his toes in the morning, and the pain worsened as the day
progressed. After three weeks of mirror box treatment twice a day, Alan no longer felt any
painful sensations from crossed toes.
The pain pathways that ascend the spinal cord
to the cortex are mirrored by complementary
descending pathways that can stimulate an
analgesic response. These pathways release
endogenous opioids at the output neurons in
the sensory area of the spinal cord. The
opioids inhibit the transmission of nociceptive
information to the cortex by the ascending
The descending analgesia pathway releases
endogenous opioids onto the output
neurons in the sensory area of the spinal
Endogenous receptors for opioids.
Receptors for opioids are found in all areas of the brain important in pain regulation. However
they are found in other areas too, explaining why the artificial opioid morphine affects many
physiological processes. Because of this, morphine is now commonly administered locally in
the spinal cord rather than systemically. This allows the dose to be decreased, which also
protects against addiction.
It is important to distinguish between a noxious stimulus and the perception of pain.
Nociceptive information is sensed in the periphery and then transmitted to the cortex by
a multi-synaptic pathway that ascends through the spinal cord. Each ascending
synapse is an important site for regulation of the response. A complementary
descending pathway modifies the input by stimulating release of analgesic peptides
Research has shown us that addiction is a disease that affects brain structure and impacts
behavior. Many of the biological and environmental factors that contribute to addiction have
been identified, and the search has begun to identify for the underlying genetic variations that
predispose and contribute to its development and progression.
Every year, abuse of illicit drugs and alcohol contributes to the death of more than 100,000
Americans, while tobacco is linked to an estimated 440,000 deaths per year.
People of all ages suffer the harmful consequences of drug abuse and addiction.
Babies exposed to legal and illegal drugs in the uterus may be born prematurely and
underweight. This drug exposure can slow the child’s intellectual development and affect
behavior later in life.
Adolescents who abuse drugs often act out, do poorly academically, and drop out of
school. They are at risk of unplanned pregnancies, violence, and infectious diseases.
Adults who abuse drugs often have
remembering, and paying attention.
They often develop poor social
behaviors, and their work performance
and personal relationships suffer.
Parents who abuse drugs often live in
chaotic, stress-filled homes with child
abuse and neglect. Such conditions
harm the wellbeing and development of
their children in the home and may set
the stage for drug abuse in the next
Pet scans indicate reduced activity
reflecting brain damage in brains of drug
What is drug addiction?
Addiction is defined as a chronic, relapsing brain disease that is characterized by compulsive
drug seeking and use, despite harmful consequences. It is considered a brain disease because
drugs change brain structure and function. These changes can be long lasting, and can lead to
the harmful behaviors seen in people who abuse drugs.
How do drugs of abuse work in the brain?
Drugs of abuse interfere with neurotransmission. For example, the synthetic opioid morphine
we discussed above mimics the endogenous opioids involved in the analgesia pathway.
Morphine is addictive because its off-target effect on dopamine synapses in the reward
pathway. In fact most drugs of abuse target this pathway directly or indirectly. Before we take a
look at that pathway in detail, let’s take a look at the characteristics of dopamine that make it
such an effective target for drugs of abuse.
Dopamine is one of three catecholamine neurotransmitters that play
important roles in brain function (the others are serotonin and
norepinephrine). They are structurally somewhat similar, but each
has a distinct receptor. Catecholamine receptors are all coupled to
G-protein second messenger modules, which means that, unlike Ach
and glutamate, they exert slow modulatory excitatory or inhibitory
effects on their target neurons. A second feature of catecholamine
neurons affects how they communicate with other axons - their
synapses are found along the entire length of the axon, not merely
restricted to the axon terminal. These en passant synapses mean
that each catecholamine neuron is able to influence a far larger area
of target cells than it would if its synapses were restricted to the
Catecholamine neurons make ‘en passant’
synapses along their entire axons. This
means they can influence a larger area of
target neurons than if the synapses were
restricted to the axon terminals.
The structure of the dopaminergic neuron means that even though dopamine pathways within
the brain are quite limited, they can nonetheless influence many distinct regions.
Dopamine pathways in the brain
There are three major dopamine pathways – 1. The
nigrostriatal system, which is involved with
movement intention, and which is damaged in
Parkinson’s disease. 2. The tubero-infundibular
system that is involved with hormone secretion and
homeostasis. 3. The mesolimbic system or reward
pathway, which is the key target for drugs of abuse.
The mesolimbic system directly or indirectly
regulates emotion, motivation, cognition, movement
and feelings of pleasure. Overstimulation of this
system, which normally rewards natural behaviors,
produces the euphoric effects sought by people who
abuse drugs and reinforces the behavior.
The reward pathway in the brain
There are three dopamine pathways in the
The dopamine reward pathway originates in a
subcortical area of the brain near the midline
called the Ventral Tegmental Area. Dopamine
neurons whose cell bodies are in the VTA end
up in the Nucleus Accumbens and the
The reward pathway connects the ventral
tegmental area (VTA) with the nucleus
accumbens (NAc). The VTA also connects
with the prefrontal cortex.
The connections between the VTA and the nucleus accumbens are called the reward pathway
because it is activated during pleasurable experiences such as eating, sex or receiving praise,
The reward pathway was discovered through the technique of intracranial self-stimulation. An
electrode implanted in different areas of the brains of rats could be activated when the rats
voluntarily pressed a lever. The rats did not regularly activate the electrode in areas of the brain
except the reward pathway: Because of the positive effects felt when this pathway is stimulated,
the behavior was reinforced.
Dopamine synapses in the reward pathway are
critically important for the effects of drugs of
abuse. The dopamine synapse is different from
the Ach synapse in how it deals with transmitter
that has been released into the synaptic cleft.
Unlike the Ach synapse that inactivates Ach with
cholinesterase, the dopamine synapse clears the
dopamine by recapturing into the presynaptic
terminal using a specific transporter.
This means that any drug inhibiting the
reuptake of dopamine from the synaptic cleft The dopamine synapse differs from the nerve –
muscle synapse – there is no inactivating enzyme
will cause dopamine effects to persist.
in the synaptic cleft.
Cocaine is one example of a drug that
blocks the function of dopamine reuptake
transporters. As a result, dopamine
levels increase in the synapse, and
consequently, the target neuron is
continuously stimulated. This constant
firing of the neurons leads to a feeling of
euphoria. In order to attain a cocaine
"high," at least 47% of the binding sites
must be blocked. In addicts, cocaine
blocks between 60 and 77%.
Compare the healthy brain at the top with the
brain of a cocaine abuser at the bottom. Why
does cocaine reduce brain activity if it increases
the activity of dopamine neurons? This is
because the output neurons from the nucleus
accumbens are inhibitory. That is to say they
shut down signaling in the regions they synapse
with. When they are stimulated by cocaine the
inhibition is even more intense. Amphetamine
has effects similar to cocaine. It also prevents
dopamine reuptake, but it also stimulates
release of dopamine into the synaptic cleft.
Cocaine reduces brain activity because output
from the nucleus accumbens is largely inhibitory.
Drugs of abuse have similar effects
All drugs of abuse increase dopamine activity at the synapse between the VTA and the NAc,
but they do it in slightly different ways:
Nicotine stimulates the VTA, where the dopamine neurons originate, directly, increasing
Heroin, other synthetic opiates (like morphine) and ethanol can also affect the VTA
synapse but they do so slightly differently: The VTA, is itself normally under significant
inhibitory control, to prevent over-activity of the reward pathway. These drugs disrupt
that inhibition. The balance
inhibition is lost and levels of
dopamine in the reward
Drugs of abuse like cocaine and
amphetamine can increase the
amount of dopamine in the synaptic
cleft between 2-10 fold more than
normal. In some cases, this occurs
almost immediately (as when drugs
are smoked or injected), and the
effects can last much longer than Many different drugs of abuse target aspects of the same
those produced by natural rewards. pathway
For instance the PET scan above was done 10 days after the patient took cocaine. The
resulting effects on the brain’s reward circuit dwarfs those produced by naturally rewarding
behaviors. This strongly motivates people to take drugs again and again.
Tolerance and Dependence
Two key consequences of drug abuse –
tolerance - increasing doses of a drug are
required to obtain the same effect - and
dependence, are both particular problems of
drugs that affect the reward pathway, because
the drive to reinforcement is so strong.
Under normal circumstances the
Bidirectional interactions between the VTA and the
reward pathway exists in homeostasis
keep the reward pathway in homeostasis. Drugs
because the communication between
of abuse increase the steady state level of the pathway.
the VTA and the NAc is bidirectional –
there are also connections between the
NAc and the VTA. The NAc communicates with the VTA by way of an endogenous
Recall that nucleus accumbens output is inhibitory. Dynorphin inhibits the VTA, and this
in turn reduces dopamine levels in the reward pathway, calming it down. Therefore,
anything that increases dopamine release at the VTA/NAc synapse will concurrently
increase dynorphin release. This will feed back onto the reward pathway, further
depressing its activity.
It is not surprising then that the drugs of abuse that increase the activity of the VTA/NAc
synapse increase dynorphin as a ‘knock-on’ effect. Down-regulation of the reward
pathway means that the same levels of drug no longer produce the same effect, leading
But because this is the reward pathway with a significant drive to reinforcement there is
pressure to continue the activity – leading to dependence.
Drugs can permanently alter the brain’s reward pathways
Dopamine’s effects are not limited to transiently depolarizing or hyperpolarizing the postsynaptic
cell. Because it acts via a second messenger it can also affect gene transcription in the
postsynaptic cell like this:
Binding of dopamine to its receptor
activates the second messenger cyclic
In turn cAMP activates the cAMPdependent kinase.
The cAMP-dependent kinase moves into
the nucleus and activates the CREB
CREB protein binds to the CRE
response element in several genes.
These genes include dynorphin and
many neuropeptide neurotransmitters.
The CREB transcription factor is activated via the
dopamine receptor pathway. It causes rapid
increases in transcription of dynorphin.
increases in feedback inhibition of the VTA we
saw before. This is a fast response.
Binding of dopamine to its receptors can also increase
other important transcription factors.
One of these is Delta FosB
Delta Fos B can activate the neurotrophic
factor BDNF (Brain derived neurotrophic
BDNF can induce the formation of dendritic
Because each spine is the site of a synapse,
inducing spines induces synapse formation.
The delta Fos transcription factor stimulates
slow and persistent changes in neuronal
structure. Neurons from the methamphetamine
abuser brain have more spines than their
In this way drugs that affect dopamine synapses can have both short-term effects on synapse
behavior (like stimulating dynorphin synthesis and release) and long-term effects on neuronal
structure like inducing synaptic spines.
Genes and addiction
While social and environmental factors
contribute to the risk of addiction, the
finding that several genes are linked to
specific addictive behaviors indicates that
there is also a genetic susceptibility.
Fortunately the reward pathway is located
in a part of the brain that is evolutionarily
very old, so that all aspects of the pathway
are almost identical in mice, rats and
humans. This means that mice, which are
a valuable genetic tool, are also useful
animal models to investigate genetic
susceptibility. Mice have the same number
of genes as humans (20,000 – 25,000) and
each mouse gene is about 85% identical to
its human counterpart (or homolog).
Among the genes now identified as
conferring susceptibility to addictive
The Dopamine D2 receptor
One particularly interesting gene whose
modifications are associated with addictive
behavior is the Dopamine D2 receptor.
Addictive behaviors are associated with defects to a
Binding of dopamine to the D2 number of genes, indicating genetic susceptibilities.
receptor inhibits cAMP formation, and
this leads to reactions that tend to inhibit addictive behaviors.
Reductions in D2 receptor levels in rats are associated with
an increase in self-administration of drugs – thought to be a
predictor of impulsive, addictive behaviors. Interestingly,
when monkeys are raised together under conditions that
result in stress in the subordinate monkey, those monkeys
have fewer D2 receptors.
The subordinate monkeys were more likely to selfadminister drugs than their dominant peers. This is
evidence that the environment can also induce molecular
changes that impact addictive behaviors.
Levels of the important dopamine D2 receptor can be modified by
stressful experiences, and then lead to impulsive behaviors.
In humans drug abuse is also associated with decrease in levels of the Dopamine D2 receptor,
indicating a feed forward inhibition at the level of the postsynaptic cell.
Summary: Addiction occurs when the activity of the reward pathway is disturbed. The reward
pathway is the mesolimbic branch of the dopamine system, in particular the dopamine synapse
between the Ventral Tegmental area and the Nucleus Accumbens. Drugs of abuse act on this
pathway in various ways to increase dopamine neurotransmission across this synapse.
Dopamine neurotransmission can lead to transient effects on the postsynaptic cell membrane,
but also effects on gene transcription. A fast acting effect on gene transcription increases
homeostasis within the pathway that may contribute to drug tolerance. Slower effects on gene
transcription result in longer acting effects on neuronal structure, including building more
synapses on the post-synaptic cell. Finally, a number of genes have been associated with
addictive behaviors. Within the dopamine pathway, decreased numbers of the D2 receptor are
associated with an increase in impulsive, addictive behaviors. Drug abuse is able to decrease
numbers of this receptor.
For lesson development
This web site has elements of a good high school Addiction curriculum: