SF - NMDA Flashcards Preview

3: Pharm of NS > SF - NMDA > Flashcards

Flashcards in SF - NMDA Deck (48):

Subunit assembly of NMDARs

3 subunit families:
* GluN1(a-h) - MUST have at least 1 (usually 2)
* GluN2(A-D) - 2nd most common
* GluN3 (A-B)

Tetramers: Diheteromeric (e.g. 2GluN1, GluN2A) or Triheteromeric (e.g. 2GluN1, 1GluN2A, 1GluN2)

Excitatory glycine receptor: 2GluN1/2GluN3 tetramer that is activated by glycine alone when co-expressed in Xenopus oocytes)

In CNS, mostly 2 GluN1 + 2 GluN2, however some areas (e.g. cortex), GluN3 is expressed

Pharmacological heterogeneity determined mainly by GluN2 (contains Glu binding site)


What is key difference between GluN1 subunits and GluN2 subunits?

GluN1 has 8 isoforms (GluN1a-h) due to ALTERNATIVE SPLICING (single gene) - varies by brain region, but not much work on the differences in pharmacology

GluN2 has 4 different genes (GluN1 A-D): subtypes from different gene (not splice variant


Topology of NMDA?


LBD (S1/S2 - clam shell)

Transmembrane region with intracellular carboxyl terminal (M1/M2/M3/M4)

M2 loop goes into membrane, forms pore


How are NMDARs activated?

Glycine binds GluN1 + glutamate binding GluN2

*glutamate: vesicles released by Ca-dependent action potentials (classic NT). glycine: not packaged into vesicles - thought to be released by glial cells? usually at saturating concentrations in the synapse so always there but receptor won’t be activated unless glutamate is released

ION FLUX will only occur when Mg²⁺ block of ion channel is removed
*often glycine + glu bind, channel opens, but nothing passes through due to Mg!


What ions pass through NMDAs?

Na⁺/Ca²⁺ in, K⁺ out

Always Ca²⁺ permeable, have slightly higher permeability than some other Glu receptors (whereas AMPAR dependent on GluR2 subunit)

Ca²⁺ permeability underlies role of NMDARs in memory learning (synaptic plasticity) and a range of neurological and neurodegenerative disorders
*over activation of receptor → cell death & neurodegeneration (AD = partly due to over-activation of NDMARs?)


How does the magnesium block work?

For Mg²⁺ block of NMDAR to occur, channel must be open i.e. glycine + Glu must be bound

If at normal resting potential (-70/-80mV); agonist bound, Mg²⁺ sits in channel pore, nothing passes through (attracted to negative charge of cell)

At ~-35mV, Mg²⁺ block is removed (Mg expelled)

*AMPARs often present at same synapse (kainate as well): activation of AMPARs depolarises membrane sufficiently to remove Mg²⁺ block of NMDARs


How is the NMDAR a co-incidence detector?

Ca²⁺ entry (NMDAR activation) is dependent on pre- and postsynaptic elements being active SIMULTANEOUSLY

Pre= Glycine/Glu release. Post = AMPAR-mediated depolarisation of post-synaptic membrane

If minimal Glu release e.g. one vesicle. probably not enough AMPAR activation to remove the Mg block. If synapse activated REPETITIVELY, depolarisation builds up, Mg block relieved + NMDARs activated.

Triggers learning of a memory & prevents too much calcium influx (triggers cell death)


What does the NMDA IV curve look like? (showing synaptic responses to NMDA)

AMPA receptors blocked, recording neuron (EPSC amplitude); & holding it at different membrane potentials

Experimenter provides depolarisation; from -104 to +36mV. At -104mV, if give stimulation to release glutamate, very little current coming through NDMA receptor (can see on IV curve)

As start to depolarise the cell, start to see more current coming through (more inward current). At -35mV, get peak current as expelling all the Mg, maximum charge going into the cell

Then start to reverse the potential; U shaped IV curve (straight line down through 0)


Physiological roles of NMDARs?

Mediate slow EPSP via Ca²⁺ + Na⁺ entry
*slower than AMPA/KARs - takes longer to reach peak & lasts longer

Ca²⁺ can activate enzymes, regulate ion channel opening & affect gene expression

Synaptic plasticity (change strength of synaptic connections + consolidate new CNS pathways)

Long term potentiation (LTP): long lasting potentiation of synaptic transmission (strength of synapse increases)

Long term depression (LTD): long lasting depression of synaptic transmission (strength of synapse decreases)

LTP + LTD may underlie learning + memory processes, disease processes & development


How is desensitisation of NMDARs affected by different subunits? Shown how?

HEK cells expressing GluN1 + GluN2: degree / timecourse of desensitisation dependent on GluN2 subtype

GluN2A: fastest & most profound desensitisation

GluN2B: slower, less profound

GluN2C/GluN2D: little or no desensitisation when activated for 1 sec by saturating (S)-glutamate and glycine

*Note difference from AMPARs: GluA1 on faster timescale so channels desensitise quicker in AMPARs


What are some pathophysiological roles of NMDARs?

EXCITOTOXICITY (excessive Ca²⁺ influx): neuronal cell death - apoptosis


PAIN transmission



Summarise different ligand binding sites on NMDARs

LBD = glutamate (GluN2) + glycine (GluN1)

UNCOMPETITIVE antagonists = block pore (like Mg)

*Ifenprodil = most studied NAM - thought to bind interface between the N-terminals (other NAM classes may bind elsewhere; i.e. between LBD + TMD - unsure about this)

PAMs = interface between the LBDs

Note: PAMs + NAMs bind same sites in AMPA and NMDARs (PAMs LBD interface, NAMs between LBD + TMD). BUT GluN2A selective NAMs can also bind at the LBD interface, therefore in NMDARs, both PAMs + NAMs can bind LBD interface


How do uncompetitive antagonists work

Use + voltage dependent:

Use = Glu + Gly bound so channel open (exposes binding site)

Voltage: blockers have +ve charge (like Mg) therefore block is voltage dependent (depolarise cell - drugs less effective)


What are examples of NMDA uncompetitive antagonists?

Phencyclidine (PCP)





What is phencyclidine?

NMDAR uncompetitive antagonist / channel blocker

Once used as GA - withdrawn as hallucinogenic, psychotic like effects (NMDAR hypofunction)


What is MK-801?

NMDAR uncompetitive antagonist / channel blocker

high affinity, adverse effects include hallucinations & memory impairment


What is ketamine?

NMDAR uncompetitive antagonist / channel blocker

Dissociative GA, induces state of sedation, immobility + analgesia- used in vets (horse tranq)

Licensed in humans for short-term pain treatment in cancer, peripheral nerve disease & spinal cord injury (blocks pain transmission particularly in spinal cord)

- still hallucinations but less severe than PCP, - - used more in children than adults; as dissociative/psychotic effects less common in children


What is memantine?

NMDAR uncompetitive antagonist / channel blocker

Well tolerated low affinity - used clinically for cognitive deficits in moderate-severe AD + the dementia of Parkinson’s

*low affinity binding but still does antagonise (without hypofunction/psychotic symptoms)
*only drug in this class licensed for dementia
*minimises excitotoxicity caused by over-activation of receptor + also normalises some of the plasticity abnormalities that occur with over-activation


What are high affinity agonists at GluN1?

8 types exist (splice variants) but no real difference in their pharmacology

Glycine and D-serine = agonists with high affinity for S1S2


Strategies for enhancing NMDAR function in schizophrenia? ** could add reading here***

Low doses of partial agonists at S1S2 of GluN1: beneficial in schizophrenia models, but beneficial effect reversed when given at higher doses: failed in clinical trials

New strategies include:

1. Developing high affinity GLYCINE TRANSPORTER (GlyT-1) INHIBITOR

(If not saturated: boost glycine? Generally thought to be saturated, but maybe decreased a bit in diseases- CONTROVERSY)

2) Development of D-AMINO ACID OXIDASE INHIBITORS (to prevent D-serine breakdown)
*not effective in human disease yet


Which disease may be associated with excitotoxicity and therefore NMDA function?

Cerebral ischaemia, Alzheimer's + Parkinsons


Why may NMDA be involved in epilepsy?

- anticonvulsant activity of NMDAR antagonists correlates with their affinity for the NMDAR (e.g. felbamate)

- prolonged discharge of AP firing, lots of glutamate release, cells depolarised (→more activation of NDMA receptors → may trigger seizures)


Why may NMDARs be involved in pain transmission

Expressed on sensory neurons,
- In hyperalgesia, antagonists can block NMDAR-dependent maladaptive plasticity (e.g. wind-up caused by NMDAR overactivation)


Why may NMDARs be involved in schizophrenia?

Linked with hypofunction

- high affinity NMDAR channel blockers cause psychotic episodes that resemble schizophrenia (NMDA antagonists create psychotic symptoms in some animals)


What are the effects of GluN1 partial agonists?

Low doses of partial agonists at the LBD of GluN1 beneficial in models of schizophrenia (potentiate NMDAR function, boosts hypofunction) but beneficial effect reversed at higher doses, failed in clinical trials


What are some GluN1 antagonists?

Kynurenic acid derivates: high affinity + selectivity for S1S2 (CNQX = glycine site antagonist)

Antagonists binding to S1S2 of GluN1 have ANTICONVULSANT, neuroprotective and ANALGSIC properties (over activation of NMDARs in pain pathways)

They have LESS PSYCHOTOMIMETIC potential that channel blockers, but pharmacokinetic properties have limited their therapeutic potential
*some animal models/basic preclinical models - but not yet sufficient for human trials


What are agonists and antagonists of GluN2?

Agonists = S-glutamate, NMDA

Antagonists = D-AP5 (R-AP5), CPP-ene, NVP, UBP141


What do we know about where GluN2 subunits are expressed?

Lots known about individual expression in diff brain regions, less about where subunits come together within same receptor (e.g. how widespread triheteromers are)

Triheteromers in the medial thalamus (GluN1/GluN2B/GluN2D) and cerebellum (GluN1/GluN2A/GluN2C)


How could NMDA function be controlled more finely?

Targeting specific GluN2 subunits or specific triheteromeric NMDAR subunit combinations

Using allosteric modulators rather than agonists/antagonists to ensure fine control

May lead to fewer adverse effects and more accurate targeting of disease processes


GluN2B - where expressed

Medial striatum + CA1 hippocampus

GluN2D: spinal cord, medial thalamus + hippocampus (greater expression in younger animals in hippocampus)
*there is an age dependent regulation for expression; GluN2D particularly - more expression in younger animals compared to adults


GluN2B - where expressed

Medial striatum + CA1 hippocampus


GluN2D - where expressed?

Spinal cord, medial thalamus + hippocampus (greater expression in younger animals in hippocampus)

*there is an age dependent regulation for expression; particularly in GluN2D - more expression in younger animals compared to adults


How can you use voltage clamp to characterise the electrophysiology of compounds on diff NMDA receptors?

*Two electrode voltage clamp

*Inject Xenopus oocyte (frog egg) with RNA that encodes for receptor of interest (e.g. GluN1 + individual GluN2 subunit) - oocyte then express receptor on the surface after a suitable incubation period

*Activation of the receptor e.g. 10µM glutamate and glycine opens the NMDAR ion channel allowing positively charged ions to flow into the cell and an inward current is recorded

*A) add agonists on their own via bath perfusion (added then sucked out) to produce a control inward current of reasonable amplitude, then continuously apply agonists and then apply increasing concentrations of an antagonist to produce a conc-response curve

*B) conc-response curves for the effect of an antagonist on four different NMDA receptor subtypes (GluN2A-D).

IC₅₀ values for the effect of the antagonist on four different NMDA receptor subtypes can be obtained and converted to Ki values (competitive antagonists)
*IC50 can be converted to measure of affinity, depending on how the experiment is done


How can you use voltage clamp to characterise the electrophysiology of compounds on diff NMDA receptors?

*Two electrode voltage clamp

*Inject Xenopus oocyte (frog egg) with RNA that encodes for receptor of interest (e.g. GluN1 + individual GluN2 subunit)

*After suitable incubation period - oocyte expresses receptor on surface

*Activate receptor e.g. 10µM glutamate and glycine - inward current recorded

1) Add agonists on their own via BATH PERFUSION (added then sucked out) to produce control inward current of reasonable amplitude

2) Continuously apply agonist, then apply increasing concs. of antagonist to produce a conc-response curves for the effect of an antagonist on four different NMDA receptor subtypes (GluN2A-D).

IC₅₀ values for effect of the antagonist on each NMDA subtypes can be converted to Ki (inhibitor constant - measure of potency of competitive antagonists)

*IC50 can be converted to measure of affinity, depending on how the experiment is done


What is the difference between NVP and UBP141?

Competitive antagonists - GluN2 subunit (binds LBD, competes with glutamate)

UBP141 about 7x more potent at GluN2D compared to GluN2B

NVP is 10x more potent at GluN2A than other subunits

These are still not very selective (NVP 10x selective, UBP141 even less selective) - therefore no way to get conc in brain that will only affect one subunit compared to the other


What is difficult about developing orthosteric antagonists of NMDA?

Very difficult to come up with orthosteric antagonists (bind LBD) that are selective between different subunits

This is because amino acid sequences in the LBD are highly conserved between the different subunits
*UBP141/NVP are the best drugs we have but still not good enough to work out exactly which receptors are where / turn into therapeutics



NAM: Selective for GluN2B containing NMDARs

Binds to ATD dimer interface between GluN1/GluN2B (Xray crystallography shows this) - thought to interfere with the function of the agonist bound LBDs

Inhibits at most 80% of response mediated by GluN1/GluN2B

*Partial inhibition by NAMs may be useful therapeutically to avoid adverse effects due to compete block of NMDAR function (NMDA hypofunction —> schizophrenia?)


What is the role of the ATD in NMDARs?

Contributes to control of ion channel open probability and deactivation rates (does not preform these functions in AMPARs)


Major difference between X-ray structure of AMPAR and NMDAR tetramers?

The LBD-ATD region is more compact in NMDARs (like hot air balloon) - drugs binding to ATD (N-terminal) in NMDAR affect things further down

This close interaction may explain why ATD contributes to ion channel function in NMDARs but not AMPARs

In AMPARs: LBD + ATD regions form minimal interactions


What is MPX-007?

More recent NAM selective for GluN2A, with a glycine concentration dependent effect

*If increase glycine conc, displaces the conc-response curve, less blocking by MPX-007 at higher glycine concs.


Where does MPX-007 bind / how does it act?

Binds at LBD dimer interface (GluN1/GluN2A): thought to push down VALINE (V783) residue on GluN2A subunit, this has knock-on effect on phenylalanine (F754) which displaces glycine from GluN1

Therefore not binding to glycine site, but pushes it out

Xray crystallography shows glycine bound structure = low affinity NAM binding site

Transition to high affinity NAM state occurs on glycine unbinding


What is GNE-6901?

GluN2A selective PAM (as shown by potentiation of Glu/Gly currents in TEVC assay of xenopus oocytes expressing GluN1/2A)

Also binds to GluN1/GluN2A dimer interface (like the NAM)

Conc.response curves show rank selectivity GluN2A > GluN2D >> GluN2C > GluN2B


Why is MPX-007 GluN2A selective?

GluN2A has valine 783 residue (which gets pushed down by the NAM)

GluN2B/GluN2C/GluN2D all have bulkier residues which inhibit NAM binding by steric occlusion


Why is GNE-6901 GluN2A selective?

GluN2A has V783, other GluN2 subunits have bulkier residues

But GNE doesn't displace the V783


Where does GNE-6901 bind / how does it act?

LBD interface (like the NAM MPX-007)

Binds tyrosine + proline residues: not interfering with glycine binding (as don't displace V783)

Exact mechanism not entirely clear but seems to slow deactivation (similar to CX614 acting at AMPARs)


How has it been shown that GNE-6901 enhances NMDA EPSPs in pyramidal neurons?

Isolated NMDAR EPSPs from hippocampal CA1 pyramidal neurons were potentiated in presence of GNE-6901

- Add agonist, then add GNE-6901: get increase in both area of NMDAR response & the peak current

*As NMDARs quite slow, tend to see more change in area of response (area underneath trace), than change on peak
- area important, will determine how much calcium comes into the cell


What is potential use of NMDAR PAMs such as GNE-6901?

*As NMDAR PAMs enhance LTP, they may find application in enhancing cognition

*NMDAR PAMs have therapeutic potential in schizophrenia (associated with NMDAR hypofunction) and cognitive disorders such as Alzheimer’s disease

- another potential smart drug through targeting glutamate system?


How has been shown that GNE-6901 enhances LTP in hippocampus?

GNE-6901 enhances NMDAR dependent LTP in the CA1 region of the hippocampus induced with one, two or three bouts of a weak induction protocol

2) Baseline response then induce LTP —> bigger response, if give GNE-6901, get even bigger response than without