Welcome back to week four of cellular mechanisms of brain function.
This week, we're going to talk about
glutamatergic excitatory synaptic transmission.
We've already studied the basics of membrane biophysics and excitability
and we've talked about synaptic transmission
as the most important way in which neurons can communicate with each other.
Now, neurons, or most neurons from the mammalian brain,
are not spontaneously active.
That is in the absence of synaptic input,
they are hyperpolarized in terms of their membrane potential,
maybe being at -70 mV,
so there are many tens of millivolts hyperpolarized
relative to the action potential threshold,
which is at around -40 mV.
In order for the nerve cells to begin to become active
to find action potentials and to participate in brain function
and the processing of information,
it's essential that these neurons become depolarized,
that they become excited, and begin to find action potentials.
The way in which neurons become excited
is through excitatory synaptic transmission,
and by far, the most important excitatory neurotransmitter
in the mammalian brain is glutamate.
Something like 80% of all the synapses in the mammalian brain
rely upon glutamate and so glutamatergic synaptic transmission
is a fundamental importance
and that's what we'll study during this week.
If an action potential occurs in a glutamatergic neuron,
that action potential will propagate down the axon
and it'll reach presynaptic specializations,
boutons, which are filled with synaptic vesicles,
each of which is packaged with glutamate,
probably at a concentration of around 100 mM of glutamate in each vesicle
and that glutamate has been put in there
specifically by the vesicular glutamate transporter
that's present on every synaptic vesicle.
Some of these synaptic vesicles will be docked and release-ready
and when the action potential invades, the bouton calcium influx occurs
does then the exocytosis of the contents of that synaptic vesicle.
The glutamate diffuses across the synaptic cleft
and there, it can bind onto postsynaptic receptors
causing excitatory postsynaptic potentials.
The fusion of a single vesicle is for it to increase
the concentration of glutamate in the synaptic cleft
to about 1 mM for a period of about 1 ms.
So we're thinking about brief, transient pulses of glutamate
as a major way in which neurons communicate with each other.
And that's what the postsynaptic cell then needs to be sensitive to.
A 1 mM puff of glutamate that then needs to act
on ligand-gated neurotransmitter receptors,
glutamate receptors, and for their glutamate receptors,
we have two major subtypes.
We have the AMPA receptor and NMDA receptor,
and we'll discuss more details about the similarities and differences
of these two ligand-gated ionotropic glutamate receptors.
Both the AMPA and NMDA receptors
are so-called ionotopic glutamate receptors.
That means that they're intrinsically ion channels
and the open probability of the ion channel
depends upon the binding of the neurotransmitter glutamate.
There are of course, transmembrane proteins
and of course, made of a sequence of amino acids,
starting from an N terminal,
running through transmembrane alpha helixes regions,
as in first transmembrane alpha helix,
there's a second alpha helix that doesn't cross the membrane all the way,
there's a third, and a fourth transmembrane helix.
Because there are only three alpha helixes
that completely cross the plasma membrane
from outside to inside,
the N terminal and the C terminal domains
are on different sides of the membrane
and the N terminal domain has a large extra cellular region
and that is also where the glutamate binding domain is encoded,
somewhere in this area circled here in red.
So the binding of glutamate changes the structure of the ion channel
and causes an increase in the open probability of the ion channel
and the ion channel itself is made from four subunits
that come together and look a bit like this.
So there are four different subunits--
come together, are sitting in the plasma membrane,
this is now a top view of the plasma membrane,
and down the middle we have the pore-forming ion channel
which then gets activated as glutamate binds
to a number of these different subunits that increases the open probability,
and we then have a cation channel that's permeable
to sodium, potassium, and in some cases, also to calcium.
There are these two major subtypes of ionotropic glutamate receptors,
the AMPA receptors and the NMDA receptors,
and they are encoded by different gene families
and so there are four genes
that are responsible for coding AMPA receptors
and there are seven genes
that are responsible for coding NMDA receptors.
When glutamate binds to an AMPA receptor,
that increases it's open probability
and the AMPA receptor is permeable to both sodium and potassium ions,
a little bit more permeable to sodium,
and it has a reversal potential of somewhere around 0 mV.
The single channel conductance is relatively small,
at around 5 pS, but that depends heavily upon the precise molecular details
of which subtypes of AMPA receptors are part of that ion channel.
NMDA receptors are also activated by glutamate.
So glutamate binds the NMDA receptor
and that increases the open probability of the NMDA receptor,
but only if co-agonists are present in the extra cellular fluid
and so either glycine or d-serine need to be present
in the extra cellular space
in order for glutamate to activate the NMDA receptor.
And usually they are,
and so usually this is not thought to be an important regulator,
but there may well be circumstances where this really works
as a co-agonist together with glutamate
in order to activate the NMDA conductance.
When activated, the NMDA receptor is permeable
to both sodium, potassium, and importantly,
is also permeable to calcium,
something that most of the AMPA receptors aren't able to do.
Calcium permeability is extremely important
because postsynaptically,
it'll turn out that calcium has important sickling properties
just as it does in the presynaptic terminal
where calcium is a final signal that causes neurotransmitter release.
Calcium postsynaptically will turn out to have an important role
in governing synaptic plasticity.
The overall reversal potential of the NMDA receptor
is also around 0 mV.
So both of the ionotropic glutamate receptors in the end,
try to bring the membrane potential close to 0 mV,
that's clearly depolarized relative to action potential threshold
and so both the AMPA and the NMDA receptors are excitatory
and they form the major excitatory neurotransmitter receptors
in the mammalian brain.
NMDA receptors have a considerably larger single channel conductance
compared to AMPA receptors.
And so they're distinguishable upon many, many features,
their ion channel permeability, single channel conductance,
and molecularly, they're of course encoded by distinct gene families.
There are also very important differences in the kinetics
of the currents that are activated.
And it's important to study the kinetics of the AMPA and NMDA receptors
under as close as possible physiological conditions.
And so, what we think might be interesting,
in terms of thinking about synaptic transmission,
are brief pulses of glutamate that last about 1 ms
and have an amplitude of around 1 mM.
Now in order to make brief pulses like this, of glutamate,
experimenters have designed an interesting way
where one has a laminar flow of two different fluids:
one fluid that contains 0 mM of glutamate,
and another one that contains 1 mM of glutamate,
otherwise containing identical concentrations
of the other components of the extra cellular fluid.
By having these two laminar flows on a piezo actuator,
one can then rapidly move this laminar flow
and then cause different concentrations
to be hitting upon a given region of space.
And so one can imagine moving this piezo actuator
and giving a 1 ms pulse of 1mM of glutamate,
to a small patch of membrane that's sitting on a patch clamp electrode.
And this membrane here,
then should preferably have glutamate receptors on it.
And glutamate receptors, we know,
of course are sitting on dendrites of neurons,
and so if we imagine a dendrite of a neuron here,
there might be glutamate receptors and synapses present
on one part of this membrane
and so let's zoom in and look at this in some more detail
and so here's the plasma membrane.
We bring in the patch electrode onto this piece of dendrite,
we can then suck a bit of the membrane
into the glass electrode.
This is our standard patch clamp recording method
invented by Neher and Sakmann,
we get the gigaohm electrical seal
of the patch of membrane around the electrode,
we can then rupture this membrane,
in the so-called whole-cell recording method
by a brief pulse of pressure, so we suck the membrane
and that ruptures this membrane.
We now have access to the inside of the membrane
through the recording electrode,
and if you now retract this electrode a little bit,
the dendrite stays in the same location,
but a bit of the membrane gets pulled out towards the membrane.
And if you keep pulling on the electrode,
eventually, this breaks
and you're left with a small patch of membrane
attached to your electrode with the outside of this membrane
being on the outside relative to your patch electrode.
And so here,
if there are glutamate receptors present,
then they have the glutamate binding site in the correct orientation.
You can give the 1 ms pulse of glutamate through the piezo transducer
and we can then measure on your patch electrode
the AMPA receptor or the NMDA receptors here,
we voltage clamp the electrode,
and we can then measure the currents
flowing through these glutamate gated--
ligand-gated ion channels.
So, if we do this for patches of membrane that contain AMPA receptors
and we give this a 1 ms pulse of glutamate
with an amplitude of 1 mM,
we find that the AMPA receptor responds very rapidly.
We get rapid activation of a glutamate gated current
that then also rapidly turns off
in response to this 1 ms pulse of glutamate.
We get, maybe, a couple of milliseconds of AMPA current.
So AMPA receptors are activated extremely quickly
and they also turn off extremely quickly.
And in fact, it turns out that it's very difficult
to have long lasting AMPA receptor currents
because even if the glutamate is prolonged,
it turns out that the AMPA current doesn't extend in time.
And there's a very strong desensitization
that takes place where, in the presence of glutamate,
the AMPA receptors desensitize
and are only able to give transient currents
lasting for a few milliseconds.
The situation for NMDA receptors is rather different.
Again, we give the same 1 mM, 1 ms pulse of glutamate,
and we activate NMDA currents,
but as you will see here,
the time scale is very different here,
we're looking at 100 ms, here we were looking at 1 ms
and in fact, it takes some time
before the NMDA current maximizes.
It maybe takes some 20 ms before the NMDA conductance
is maximally opened,
long after the glutamate has already gone.
And the NMDA receptor currents remain active
for hundreds of milliseconds after the glutamate pulse.
And so it's clear that the temporal kinetics of AMPA conductance
and NMDA conductance are extremely different,
with one or two orders of magnitudes slower activity in NMDA receptors,
compared to the extremely fast AMPA receptors.
There are also important differences in the voltage dependence
of the currents generated by AMPA and NMDA receptors.
AMPA receptors are basically voltage insensitive,
in at least, most of the forms of AMPA receptors.
And so if we think about the conductance activated by a pulse of glutamate,
this is independent to voltage,
and that then gives rise to a linear current voltage relationship.
For NMDA receptors, the situation's rather different.
At hyperpolarized potentials,
and that's something hyperpolarize theta - 40 mV,
the conductance of the NMDA receptor is strongly reduced.
The open probability that the amount of current flow
is blocked here at negative potentials,
and at positive potentials, we come here with a full activation.
And so we get these current-voltage relationships
that look very similar to what we saw before
for voltage-gated sodium, and voltage-gated potassium,
and calcium conductances,
but here, the reversal potential here is close to 0 mV,
of course in both cases, for the AMPA and the NMDA receptor,
whereas for the voltage-gated sodium conductors,
we of course, had reversal and voltage-gated
at the sodium reversal potential.
So there's an obvious difference here under physiological conditions
between the non-voltage dependence of AMPA receptors
and a strong voltage dependence for NMDA receptors.
And although this looks very similar to the phenomenology
of voltage-gated sodium channels,
in fact, the physical basis is very different.
In fact, the NMDA receptors aren't intrinsically voltage sensitive,
but they gain that attribute because of a magnesium block.
And so in the presence of zero magnesium in the extra cellular space,
in fact, the NMDA receptor is linear--
voltage and current--
and is only in the presence of extra cellular magnesium
that we have a strong dependence upon voltage
for the NMDA receptor.
Now, physiologically, magnesium is always present
in the extra cellular solution.
And so the voltage dependent magnesium block,
or the NMDA receptors is extremely relevant,
functionally to the properties of the NMDA receptor.
The fact that the block only occurs at negative potentials
is because that's when there's an electric field
that tries to suck magnesium inside the ion channel
and the magnesium is in fact, blocking the ion channel
by sitting in the ion channel,
binding to a relatively high affinity site
and preventing the permeation of other ions.
And so the magnesium pops in and out
in a so-called fast, flickering block where it directly blocks the permeation
of the other ions and doesn't affect the open probability
of the ion channel, but is just simply being blocked
by the presence or absence of magnesium
and normally with 1 mM magnesium present in the extra cellular space
and negative membrane potentials that confers a very strong block
on the conduction through the NMDA receptor,
and that's why we have this interesting shape
where at negative potentials, magnesium comes in,
blocks current flow through the NMDA receptor,
and as we get to more and more positive potentials,
then the magnesium gets pushed out from this internal binding site
and we get the full current flow of the other ions
through the NMDA receptor.
So there's not a voltage sensing domain of the NMDA receptor,
but rather a voltage dependent block by magnesium
in the pore of the ion channel.
So we've seen that there are two
very different ionotropic glutamate receptors
that are present in the mammalian nervous system.
We have the AMPA receptors that are extremely fast,
giving rise to millisecond duration currents
in response to a 1 ms pulse of glutamate
and the AMPA receptors don't depend upon the membrane potential.
So at -60 mV or +60 mV, the amount of current flow is the same
and the reversal potential is somewhere around 0 mV.
For the NMDA receptor, the situation's really, very different.
We have long lasting, slow currents
in response to just millisecond pulses of glutamate.
It takes tens of milliseconds for the NMDA receptor to open
and once the glutamate is bound, it remains bound
and their ion conductance can remain active
for hundreds of milliseconds after the glutamate has already long gone
from the synaptic cleft.
The NMDA currents are also interesting
because they're actually blocked at resting membrane potentials.
So at the resting membrane potential of a neuron,
at around -70 or -60 mV,
if the glutamate is puffed onto NMDA receptors,
there's actually very little current flow
because magnesium is blocking that ion channel.
So, NMDA receptors depend upon both postsynaptic depolarization
and the presence of glutamate,
whereas AMPA receptors are always active if glutamate is present.
There's considerable diversity
amongst the different types of AMPA receptors,
and there are in fact, four genes that encode AMPA receptors
in the mammalian genome,
and these are then labeled in their gene names
as glutamate receptor ionotropic ampa
and then subtypes 1, 2, 3, and 4.
So that's the gene names that then give rise
to the different types of subunits
that are typically labeled GluA1, GluA2, 3, and 4.
And in older papers, you will read a different nomenclature,
whether it's called GluR1-4 or GluRA-D.
So there's a variety of different nomenclatures
and this is the most modern one that should be used from now on.
Now, most AMPA receptors contain the so-called GluA2 subunit.
And if the AMPA receptor contains the GluA2 subunit,
then as described,
we have linear current-voltage relationships,
and there's no calcium permeability.
Now, some AMPA receptors don't have the GluA2 subunit.
And under those conditions, the current-voltage relationship
is really quite different.
So now we're thinking about GluA2-lacking AMPA receptors
and they have a so-called inward rectification
where they like current to go into the cell and depolarize it,
but they don't like current to go out of the cell,
in fact, there's an intracellular polyamine block
at positive membrane potentials.
In addition, GluA2-lacking AMPA receptors are permeable to calcium,
and so have interesting sickling roles.
And so whereas most AMPA receptors
are composed of combinations of GluA2 with GluA1
or GluA3 heteromers,
there are some which don't contain the GluA2,
and are simply GluA1 or 3, for example,
as a subunit they lack the GluA2
and then they have these interesting voltage dependencies
that again, depend upon intracellular components
like polyamines that block the ion channel
from the inside.
So there's some diversity in the AMPA receptors
and this is then further regulated
and involved in different types of synaptic plasticity
and other sickling features.
In addition to AMPA receptors,
there are also Kainate receptors that bear some resemblance
to the AMPA receptors, but they have a much less defined function.
And there are five genes
that are encoded by the so-called kainate receptor family--
glutamate receptor ionotropic kainate 1-5,
GluK1-5, they used to be labeled GluR5-7; KA1, 2.
They are ionotropic ligand-gated glutamate receptors
just like the AMPA receptors to which they're more similar
compared to, at least, the NMDA receptors.
They seem to have smaller currents,
they seem to be slower,
in some cases, they may also be located on presynaptic specializations
rather than postsynaptically,
but on the whole, much less is known
about the function of the kainate receptors.
The NMDA receptors form the other major family
of ligand-gated glutamate receptors,
and they're composed of seven different genes.
There's the grin1 gene, grin2A-D, and grin3A, and B.
And what's important about the NMDA receptors
is that the NR1 subunit, the GluN1 subunit,
is absolutely required for function.
So all NMDA receptors are thought to have the GluN1 subunit
as part of the four subunits.
So one of these will be the NR1 subunit, at least.
This will then combine with other subunits,
and the most common combination
is with NR1 with 2A
or NR1 with 2B.
And that's most of the glutamate receptors
in the mammalian brain are encoded
by either the NR1 2A or the NR1 2B.
Both of these combinations are very strongly blocked by magnesium
as we've already seen, they have an almost complete block
at, let's say, at around -80 mV,
the NMDA receptor's completely blocked by the presence of magnesium.
However, two other subunits, the 2C and the 2D,
have a much weaker block
with respect to magnesium.
And so in comparison, they would have something this,
where there's a substantial amount of unblocked NMDA receptor current
even at these very hyperpolarized membrane potentials.
So there's differences in the degree of magnesium block
comparing across the different subunits of the NMDA receptor.
We'll also discuss the very long time constants
where in response to a 1 ms pulse of glutamate,
the 2A subunit might have a time course of something like 100 ms,
whereas the slowest of the subunits might have times of around 1 s
for it's sort of half decay time,
and the 2B and 2C are somewhere in the middle.
And so the time constants of the conductance activated by glutamate
is extremely different comparing across these different subunit combinations.
And so you can get a great deal of diversity
in NMDA receptor function by having different combinations
of these different NMDA receptor subunits present at the synapse,
and indeed different cells and different synapses
have very different features.
Finally the GluN3, subunits A and B,
have less known functions, but might function presynaptically
or play other roles in sickling and synaptic plasticity.
Finally, of course, all though we're primarily interested
in the fastest forms of communication
between neurons mediated by the ionotropic glutamate receptors
that generate postsynaptic potentials on the millisecond time scale,
we mustn't forget about the metabotropic glutamate receptors.
These are 7 transmembrane receptors,
they couple through g-proteins,
activate [inaudible] through sickling pathways,
like phospholipase or they inhibit adenylate cyclase
in track of calcium sickling.
They probably have multiple, diverse roles,
but are yet to be fully explored.
And there's eight genes that encode these metabotropic glutamate receptors
and these are likely to have slower sickling functions
than the ionotropic glutamate receptors
that are working on the millisecond time scale.
So the take home message from today's lesson,
is that there are two major subtypes of ionotropic glutamate receptors.
There are the AMPA receptors,
that are responsible for the millisecond activation
of a synaptic conductance in response to glutamate,
and there are the NMDA receptors,
that are working on the tens of milliseconds
or hundreds of milliseconds time scale.
They're activated more slowly,
they last for hundreds of milliseconds,
long outlasting that pulse of glutamate from the presynaptic specialization.
The conductances that are opened,
are both excitatory for AMPA and NMDA receptors,
they reverse at 0 mV,
the AMPA conductance being permeable to sodium and potassium,
and the NMDA conductance, in addition, allowing calcium entry
into the postsynaptic specialization,
where it'll turn out it has an important role
in controlling synaptic plasticity.
In addition, the NMDA receptor has a very important dependence
upon the postsynaptic membrane potential.
At negative potentials, the extra cellular magnesium
enters inside the pore of the NMDA receptor and blocks it,
making the other ions impossible for them to permeate.
And so through this voltage dependent magnesium block,
the NMDA receptor is both sensitive
to the state of the postsynaptic cell,
it needs to be depolarized in order to unblock the magnesium,
and it, of course, also needs to have glutamate
release from it's presynaptic terminal.
So the NMDA receptor has an interesting coincidence role,
where it's both sensitive to the activity of the presynaptic cell,
as well as a postsynaptic cell in which it's embedded.
And we're going to learn more about the importance of that coincidence detection
as we consider synaptic plasticity in the next lesson.