Synaptic Plasticity and Memory

Question 1

Describe the taxonomy of memory

Answer 1

Declarative (explicit); semantic and episodic memory (hippocampus, medial temporal lobe)

Implicit memory;
Skills and habits (striatum, motor cortex, cerebellum)
Priming (neocortex)
Basic Associative Learning (emotional; amygdala, skeletal musculature; cerebellum)
Non-associative learning (reflex pathways)

Question 2

What is the EPSP a measure of?

Answer 2

Strength of AMPAR mediated synaptic transmission

Question 3

What are the 3 forms of synaptic plasticity

Answer 3

LTP
LTD “de novo”
LTP depotentiation

Question 4

What are the 3 classical properties of LTP that may be relevant to its role in memory?

Answer 4

Persistence; a single HFS (tetanus) can induce LTP for hours/weeks

Input specificity; when you tetanize a group of afferents, only the tetanized ones are potentiated. Important for memory storage (only want synapses that participate in memory to be potentiated)

Associativity; pair stimulation of a weak input to a strong input; the weak input is potentiated (cells that fire together wire together). Extension to this is cooperativity

Question 5

Is associativity relevant to all forms of LTP?

Answer 5

No, but is relevant to NMDAR dependent LTP in hippocampus

NMDAR detect a coincidence of post-synaptic depolarization and pre-synaptic glutamate response

Question 6

Describe a behavioural impact of associative conditioning?

Answer 6

Classical conditioning;

Neutral stimulus (bell) rang = conditioned stimulus 
Unconditioned stimulus such as food = salivation 

When bell and food given at the same time there is an unconditioned response of salivation due to presentation of food
After this training; when a bell is rung, there is salivation as a conditioned response

Question 7

Describe Hebb’s postulate about associative conditioning

Answer 7

When an axon of cell A is near enough to excite a cell B and repeatedly or persistently take part in firing, some growth process or metabolic change takes place in one or both cells such that A’s efficacy as one of the cells firing B, is increased

Question 8

Is it feasible to make determinations about behaviour based on the cellular change that occurs?

Answer 8

No; the role of synaptic plasticity in a particular form of memory is determined by the neural circuit which is implicated

E.g.; a reflex network vs a distributed associative network

Question 9

What are the criteria for assessing and memory hypothesis?

Answer 9

Anterograde alteration; blocking changes in synaptic strength should block memory formation

Retrograde alteration; changing of synapses after the learning task should wipe out the memory that is stored

Detectability; if synaptic changes happen during learning they should be detectable

Mimicry; selective changes in the brain can engineer memories for something that didn’t actually happen

Question 10

Describe anterograde alteration

Answer 10

Interventions that prevent or limit the induction of synaptic weight change during a learning experience should block or impair the animals memory of that experience.

Interventions enhancing plasticity might also improve learning

Question 11

What are the issues with attempting to show anterograde alteration?

Answer 11

What physiological, pharma or molecular genetic manipulations to use? Side-effects?
How to ensure appropriate regional specificity of the treatment for the type of learning/ memory under investigation
Inducibility and reversibility of the treatment?

Question 12

Describe the system for studying spatial memory

Answer 12

Morris Water Maze

Question 13

Does NMDAR blockade impair spatial memory?

Answer 13

Yes; NMDAR blockade (D,L - AP5) impairs water maze reference in memory formation

LTP was blocked relative to the vehicle treated controls

Question 14

How can intraventricular drug delivery be improved?

Answer 14

Intrahippocampal drug delivery via catheter

Fairly selective due to hippocampus surrounded by white matter

Question 15

Describe D-AP5 on the ability of mice to on day learning

Answer 15

Delay-dependent impairment in a delayed matching-to-place task
Only long term memory affected (short term memory was normal)
NMDA receptor activation is required to form a long term memory

Question 16

Describe the role of hippocampal NMDA and AMPAR in the encoding and retrieval of one-place memory

Answer 16

NMDA antagonist; AP5
Blocking NMDA hippocampal receptors for encoding completely blocked the correct exploration
However, AP5 injected before retrieval had no problems with retrieval of memory

AMPAR antagonist (CNQX);

Blockade of AMPARs before retrieval drastically reduced the correct exploration (retrieval of memory)
This links to the fact that AMPARs mediate fast synaptic transmission - functional lesion. Memory can no longer be expressed

NMDA receptors are not required for retrieval of memory, however they are required to encode memories

Question 17

Describe the role of GluA2 lacking AMPARs in synaptic plasticity and memory

Answer 17

GluA2 lacking subunits are permeable to calcium
LTP inducing

LTP induction selectively involved the insertion of GluA2 lacking AMPARs in post-synaptic membrane

This trafficking depends on C-terminal region of GluA1 regions

C-tail of GluA2 is crucial for removal of Glu2 from synapse = LTD

Question 18

Describe Zhou et al. work on gluA2 lacking AMPARs and LTP

Answer 18

Necessary for Ca1 LTP

Genetically modified GluA1 receptors to contain the gluA2 C-tail.
This led to a complete block of LTP
Without this GluA1 C tail; there is a reduction in AMPAR trafficking to post-synaptic membrane

Furthermore, impact on behaviour; necessary for spatial memory formation (Morris Water Maze)

Question 19

Summarise NMDARs and AMPARs in memory encoding and retrieval

Answer 19

For place memory, hippocampal NMDAR activation with the subsequent upreg of GluA2 lacking AMPARs is necessary for the encoding of memory

Further NMDAR activation is NOT required for memory retrieval.

Pharma blockade of AMPARs with CNQX prevents successful memory retrieval

Question 20

Is it possible to saturate LTP physiologically on the extrinsic or intrinsic pathways of the hippocampus?

Answer 20

Easiest pathway to observe is EC - DG (perforant pathway)

Hypothesized that saturation of LTP in the perforant pathway should inhibit hippocampal dependent learning

Question 21

How can LTP saturation be achieved?

Answer 21

High frequency stimulation across the fibre bundle using bipolar electrodes with a collar and tip (cross bundle tetanisation)

Attempt to potentiate as many inputs as possible

3rd electrode in central of perforant path to induce LTP et end of task to determine if saturation was effective

Question 22

What were the results of the LTP saturation in the perforant pathway?

Answer 22

Water Maze; saturation of dentate LTP prevents learning of spatial information

Question 23

Describe the impact of targeted deletion of NMDA receptors in the CA3 region

Answer 23

Mice with a targeted deletion of NMDA receptors in CA3 trained in watermaze reference memory task

Retention was normal with all cues present, but impaired when there was only a limited set of cues available

Place cells were more diffuse in CA1 with a limited set of cues, but normal with a full set

Question 24

Is the CA3 required for pattern completion?

Answer 24

NMDAR plasticity at CA3 recurrent synapses is necessary for encoding to occur in such a way that pattern completion can occur with partial information

Reactivation of previously strengthened recurrent connections is unnecessary for recall under full-cue conditions

Question 25

What is meant by retrograde alteraion?

Answer 25

Interventions that alter the spatial distribution of synaptic weights induced by prior learning should alter the animal’s memory of that experience

Question 26

What are the issues with demonstration of retrograde alteration?

Answer 26

What physiological, pharma or molecular genetic manipulations should be used?
What are the side effects/
How can you ensure that the appropriate regional specificity of the treatment is relevant to the type of learning/ memory under investigation
How soon after the prior learning experience should the animals be manipulated?

Question 27

How can existing memory traces be disrupted?

Answer 27

LTP induction

After LTP induction, a normal sparse pattern of learning-induced potentiation (A) is unreadable against a background of artificial synaptic changes induced after training that contain no meaningful information

Question 28

Describe a trial that demonstrates the disruption of memory via LTP induction

Answer 28

Brun et al

Trained in water maze

HFS given with a saline vehicle = significant reduction in time spent in correct one of water maze

HFS given with CPP (NMDAR antagonist) = no reduction in time spent in the correct zone of the water maze; spatial memory not impacted

LFS with CPP or sal; no impact on retrieval/ retrograde memory

Question 29

Describe PKM-zeta

Answer 29

Isoform of PKC; lacks the regulatory component, only has catalytic component
Constitutively active
Involved with continuous trafficking of AMPARs to result in a long lasting increase in synaptic strength

Question 30

Can PKM-zeta be blocked?

Answer 30

Yes; ZIP

Question 31

What is the effect of ZIP on LTP?

Answer 31

ZIP (PKM-zeta inhibitor) after LTP with completely reverse LTP back to baseline levels

However, ZIP doesn’t impact the baseline

ZIP “erases” spatial memory
Therefore, reversal of LTP acts to erase memory

Question 32

What is detectability?

Answer 32

If an animal displays memory of previous experiences, change in synaptic efficacy should be detectable somewhere in the nervous system

Question 33

What are the issues with dectability?

Answer 33

Where should you look in the brain (memory storage)
How big a change is expected (storage capacity)
Will induction preclude further induction? (saturation)

Question 34

Is there an impact of exploration behaviour and brain temperature?

Answer 34

Yes; increasing EPSP slope results in an increase in brain temperature

Question 35

How can the role of the amygdala in memory be studied?

Answer 35

Fear conditioning
Behavioural LTP induced by fear conditioning

Behavioural LTP is induced in the lateral nucleus of the amygdala by pavlovlan fear conditioning.

Auditory fear conditioning does result in an increase in synaptic strength

Question 36

Does fear conditioning change AMPAR expression?

Answer 36

Similar to LTP; AMPAR is added to post-synaptic membrane during fear conditioning

Virus mediated overexpression of GFP tagged GluA1 AMPARs in lateral amygdala
Subunits from homomeric receptors containing only GluA1 subunits

Most normal AMPARs contain a mixture of subunits, including both GluA2
GluA2 containing receptors pass ionic currents almost equally well in both directions (i.e. at -60mV and +40 mV)

Homomeric GluA1 receptors pass much less current in the outward direction (greater rectification). This does not affect normal physiological function, but provides a marker for AMPA insertion

Fear conditioning, similar to LTP, results in an increased rectification, indicating that new synaptic AMPARs are inserted during memory formation

Question 37

Does skill learning have an impact on plasticity?

Answer 37

Skill learning causes an increase in evoked responses (EPSP) recorded in the motor cortex (CI)
This was a contralateral effect

Skill learning results in a partial occlusion of LTP and a facilitation of LTD. Skill learning shifts LTP induction upwards

The thing we don’t know; is how this information is stored or retrieved

Question 38

Describe behavioural familiarity to a visual stimulus

Answer 38

The more the animal is familiar to the visual stimulus, the less it moves

Visual recognition memory formation causes synaptic changes - visual evoked response grows with increased stimulus with a decrease in movement (acquired visual recognition memory)

Question 39

Describe the learning related changes in the hippocampus

Answer 39

Step-through inhibitory avoidance

Inhibitory avoidance training induces hippocampal LTP. Increase in synaptic strength

Question 40

Describe a study that combines detectability and retrograde alteration

Answer 40

Development of an optoprobe to label dendritic spines enlarged after motor learning

Mice trained to walk on a rotarod without falling off. This training induced dendritic spine enlargement and de novo spine formation - increased in fEPSP in motor cortex after learning

Photoactivation of optoprobe to shrink spines (PaRac1) resulted in the loss of memory of the recently learned motor task

Question 41

Describe Arc (IEG)

Answer 41

Expressed when synapses are active

RNA is targeted to active dendritic spines

Question 42

Describe PSD-95

Answer 42

Expressed in post synaptic density (area with proteins required for synaptic plasticity and signal transduction)

Question 43

How are the inputs form the auditory cortex to amygdala organised?

Answer 43

Tonotopically; organised by sound frequency
Sound of one frequency will activate a set of synapses
Different frequency will activate a different set of synapses

Question 44

Describe c-fos

Answer 44

Only expressed in activated neuronal cells

Question 45

Describe the study that demonstrated synapse-specific encoding and erasure of fear memory in the amygdala

Answer 45

Knock in mice in which Cre is driven by c-fos expression (i.e. only active cells)

Injected with a Cre-dependent adenovirus construct that drives yellow fluorescent protein (YFP) expression; labels active inputs during fear conditioning (CS+ or CS-)

Replacement of YFP with ChR2 construct enables active afferents to be stimulated ex vivo by blue light

Ex vivo stim of CS+ but not CS- elicits larger AMPA/NMAR EPSC ratio in mice trained to assoc CS+ with foot-shock relative to untrained controls

LFS depotentiation optogenetic stim of CS+ axons reduced the CS elicited freezing - i.e. reversed the fear memory

Question 46

Describe mimicry

Answer 46

If it is possible to induce an appropriate spatial pattern of synaptic weight changes, the animal should display apparent memory for some past experience which did not occur in practice

Formation of artificial memories

Question 47

Is it easier to achieve mimicry in some circuits in comparison with others?

Answer 47

Easier in reflex network leading to a straightforward change in behaviour - e.g. amygdala for fear conditioning

More difficult in distributed associative network (e.g. CA3)