Synaptic Plasticity Flashcards
Who first suggested synaptic plasticity?
DONALD HEBB: when 2 cells active at same time, synapses between them get stronger
Hebbian theory often summarised as “cells that fire together wire together” - however he emphasised that cell A must fire just before cell B - this foreshadowed what is now known about spike-timing-dependent plasticity, which requires temporal precedence
- taken from Wikipedia
Info is stored partly through alteration in the synaptic efficiency of synaptic transmission ~100 billion brain cells in brain, each connected to ~1,000 others = 1000 trillion connections
What did Brenda Milner discover?
Identified hippocampus as a structure important for learning & memory - studied H.M. who had both hippocampi removed for intractable epilepsy; unable to store any new information
- learnt skill e.g. piano / ping pong but no conscious recollection of learning the information
Who discovered LTP? How (methods)
Bliss + Lomo (1973): electrode into dendate gyrus in anaesthetised rabbits
Stimulated PERFORANT path (afferent inputs cortex→ granule cells) and recorded OUTSIDE cell to sample local field potentials (because recording for long periods; therefore difficult to record from individual cells)
*As hippocampus is highly organised, can get large stable field potential recordings - indication of excitatory synaptic transmission synapses from entorhinal cortex→ granule cells of dentate gyrus
What is local field potential?
From Wikipedia:
An electrophysiological signal generated by the summed electric current flowing from MULTIPLE nearby neurons within a small volume of nervous tissue (typically in anaesthetised animal or thin in vitro slice)
- Voltage produced across local EC space by APs + graded potentials in local neurons
What did Bliss and Lomo (2010) find?
Rabbit dendate gyrus - first example of long-lasting increase in synaptic efficiency in CNS
If stimulate at low frequency (e.g. every minute), response would stay fairly constant (still a bit of intrinsic biological variability)
Brief period of high frequency stimulation (tetanus): response gets bigger (potentiates)
*long-term / long lasting: 4 periods of tetanic stimulation, = elevatedresponse elevated several hours later
Parallel experiment: conscious, freely moving rabbits, induced LTP that lasted weeks
What are three key properties of LTP?
Input specificity
Co-operativity
Associativity
What is input specificity?
If activate 2 sets of fibres that impinge on same cell / group of cells, and induce LTP in one set, there is no change in the other set (synapse will strengthen, but rest of synapses on that cell will not change)
*Therefore, unit of modification is not at neuron level (~100billion) but at synapse level (~100 trillion! brain processing power due to synapses)
What is co-operativity?
Weak input which activates just a small number of fibres will not induce LTP
Must activate several fibres simultaneously; imparts an intensity threshold to the process
i.e. won’t store all info every time a synapse is activated; must be important info which is signalled by many synapses that are active at same time
What is associativity?
Strong input can help a weak input to potentiate, if it is active around the same time
- basis of associative learning/memory
- associativity can be provided by other inputs, doesn’t have to be same type of synapse, could be neuromodulatory substance e.g. ACh, DA, 5HT
What is the Schaffer collateral pathway?
Connection from CA3 pyramidal cells → CA1 pyramidal cells (where most work done, but principles 1st discovered here generally apply throughout CNS)
AP → VGCCs open → glu vesicles released → brief [Glu] increase in cleft → diffuses + activates post-syn receptors → net flow positive into cell (depolarisation e.g. from -70 to -50) → EPSP → channels close, potential back down (*intracellular micro-electrode: record fluctuations in membrane potential)
What transmission do AMPArs mediate? How do we know this?
Fast synaptic transmission
Blake et al (1988):
CNQX completely blocks the EPSP (induced by LOW frequency stimulation of the pathway i.e. if stimulate ~30s, record EPSP that is blocked by CNQX)
Therefore, the receptors mediating fast synaptic transmission at this synapse = AMPARs
What is the involvement of NMDA receptors during low frequency transmission? How do we know?
No involvement (Heron et al 1986)
- No effect when low-frequency stimulation performed on NMDA receptors (D-AP5 made no difference to the EPSP)
- Most likely explanation: NMDARs are present at the synapse, because applying NMDA locally will excite the cells, but not contributing to synaptic response that is induced by low-frequency stimulation
What did Collingridge et al (1983) show?
NMDARs don’t mediate low-frequency transmission, but trigger LTP
- AP5: highly NMDAR specific + blocks LTP induction
- First done at Ca1-Ca3; now know from additional experiments that LTP in most synapses requires NMDAR activation
- These experiments have been looking at LOW frequency stimulation
What did Herron et al (1986) show? (high-frequency stimulations)
With high-frequency stimulation (100Hz at 10ms intervals): AP5 now makes a big difference
With AP5: 20 AMPAR-mediated EPSPs
Wash out AP5 = Get APs as there is an additional slow component
Subtract AP5 trace from wash out trace (APs): see slow depolarising response (mediated by activation of NMDARs). Slow depolarisation (NMDARs) summated with AMPAR-mediated depolarisations, causing cell to fire APs.
Do NMDARs contribute to synaptic transmission?
Yes, but in a frequency-dependent manner
*High-frequency discharges activate NMDARs, low-frequency discharges do not activate NMDARs to same extent
Cannot say that NMDARs are involved in plasticity but not transmission: they are involved in high-frequency transmission (a major form of communication in our brains)
What did Ault et al (1980) show?
Hemisected frog spinal cord (lacks Mg) + measured responses to diff agonists
- Applied Mg to block synaptic transmission + incidentally found that Mg blocked depolarisations of neurons, specifically to NMDA (Kainate depolarisaitons were spared)
- can detect effects with as little as 10uM of Mg, in human brains; physiological endogenous [Mg] ~1mM; so 90%+ inhibition of NMDA receptors
What did Herron et al (1985) show? (low frequency stimulations)
Performed low frequency stimulation under normal conditions (hippocampus has Mg in normal conditions); then remove Mg
Response massively larger without Mg: lots of AP firing which is very sensitive to AP5
Conclusion: no NMDAR activation in physiological Mg during low-frequency stimulation. In absence of Mg, responses substantially enhanced - large component of this is mediated by NMDARs
*Important to have Mg as this amount of excitation is epileptiform! massive excitation of NMDARs → seizures & cell death
What did Nowak et al (1984) show?
Patch-clamp single-channel recording from cultured striatal neurons (cell surface electrode, isolate small piece of membrane, record current from single receptor)
At -60mV (close to resting) with Mg: flickering block of channel, at positive potentials: no block whatsoever, if hyperpolarised: block intensified (-70mV, block about 90%)
Mg²⁺ blocks NDMAR channels in extremely voltage-dependent manner
Limitation: applies to cultured neurons with exogenously applied agonist
What did Collingridge et al (1988) show?
Voltage-clamped cell: measured EPSCs (in unclamped cell, would cause EPSPs)
Physiological conc. of Mg (~1mM) + depolarising to point where Mg block predicted to be low. Applied DAP5.
NMDA = much slower activation kinetics, longer lasting (compared to AMPA)
Conclusion: High-frequency transmission causes depolarisation of the post-synaptic membrane, partly by temporal summation of AMPAR-mediated responses (positive charge builds up - Mg block repelled) and NMDAR activation (NMDA component builds up slowly because of the activation kinetics of the NMDARs)
*But.. can’t just be summations of excitation, when excitation occurs you always have inhibition as well (if excitation-inhibition added together, inhibition would always be stronger..)
What did Collingridge et al (1988) show?
Voltage-clamped cell: measured EPSCs (in unclamped cell, would cause EPSPs)
Physiological conc. of Mg (~1mM) + depolarising to point where Mg block predicted to be low. Applied DAP5.
NMDA = much slower activation kinetics, longer lasting (compared to AMPA)
Conclusion: High-frequency transmission causes depolarisation of the post-synaptic membrane, partly by temporal summation of AMPAR-mediated responses (positive charge builds up - Mg block repelled) and NMDAR activation (NMDA component builds up slowly because of the activation kinetics of the NMDARs)
In high-frequency transmission, why can’t the depolarisation of the post-synaptic membrane be purely due to summations of excitation?
When excitation occurs you always have inhibition as well (if excitation-inhibition added together, inhibition would always be stronger..)
Lots of inhibitory inter-neurons types exist, when EPSP induced, also induce IPSP by activating these neurons (usually stimulation of excitatory fibres will also activate some inhibitory interneurons - synaptic delay where inhibition delayed by 1-2ms)
How are AMPARs inhibited after depolarisation?
Shortly after start of AMPA response, activation of GABA response via GABA-A receptors (chloride influx)
- This hyperpolarises cell & INTENSIFIES Mg block of the NMDA receptors
- GABA inhibition is very powerful, not only hyperpolarises membrane but also affects conductance (more difficult for cell to depolarise)
What did Collingridge et al (1998) show about GABA inhibition?
In presence of Mg, blocked GABA inhibition with picrotoxin
Low-frequency stimulation: part of the response was mediated by NMDARs
*response becomes slower, because blocking inhibition causes part of the response to be caused by NMDA
Conclusion: GABA inhibition has critical role of hyper-polarising cell to intensify Mg block under normal, low-frequency conditions
Therefore picrotoxin has similar effect to removing Mg (convulsions)
Outline the different components of the synaptic response (including inhibitory regulation)
Fast AMPA, fast GABA-A (chloride influx), slow GABA-B (potassium efflux), slow NMDA component (only manifests depending on membrane potential of cell, due to Mg block)
Net result: fast EPSP, curtailed by IPSP (prolonged by GABA-B response), due to hyperpolarisation + very little activation of NMDARs during low-frequency transmission
Why is GABA-A fast and GABA-B slow?
When GABA-A activated, after a short delay, also activate GABA-B receptors (B = GPCRs, therefore small delay to activation as G proteins dissociate then K+ channel activated)
What did Davies et al (1991) show? (paired pulse)
Activated GABAergic interneurons whilst blocking glutamate excitation (e.g. CNQX / DAP5) - recorded IPSCs (outward currents)
2 stimuli (separated by 200ms) - 2nd response much smaller than the first
Less outward current in response to 2nd pulse: when GABA released, some binds pre-synaptic autoreceptor: functions to inhibit subsequent GABA release
Therefore, DEPRESSION of synaptic INHIBITION during HIGH-FREQUENCY transmission
If apply GABA-B antagonist - paired pulse depression is BLOCKED therefore AUTORECEPTOR = GABA-B
What did Davies et al (1991) show? (priming)
Used 5Hz interval: mimics theta rhythm
1 stimulus, then 4 stimuli 200ms later
1st response depresses GABA release, then when only 4 stimuli are delivered at optimal depression: substantial inhibition of GABA enables the 4 stimuli to activate NMDARs
This is due to GABA-B autoreceptor activity, as GABA-B antagonist reduces the EPSP
What is priming / tetanus?
Priming = stimulate with one pulse, then 200ms later, 4 stimuli at 100Hz - induces LTP very efficiently
LTP classically induced by tetanus -prolonged high-frequency stimulation (e.g. 100 shocks at 100Hz)
BUT LTP can be induced by just a few stimuli if correct timing, optimal timing = THETA rhythm (rhythm during exploratory activity -likely rhythm when we are leaning things)
What are the steps leading to LTP during HIGH-frequency stimulation?
Activation of AMPARs
Depression of GABA inhibition (via GABA-B autoreceptor)
Weakened inhibition enables depolarisation of excitatory synapse - Mg block then repelled from NMDARs
NMDARs contribute to synaptic response
*this means 4 stimuli are sufficient to induce LTP
What did Alford et al (1993) show? (calcium)
CA1 neurons; rat hippocampal slices, whole patch clamp + calcium imaging
High frequency stimulation leads to synaptic activation of NMDARs & generates local Ca2+ signal in spines
Once activated, permeable to sodium ions, adds depolarisation - POSITIVE feedback: further NMDAR activation: ~1/10 ions through receptor is calcium
DAP5 - blocks the depolarisation + blocks calcium
Ryanodine - (intracellular calcium channel blocker) - does not block the depolarisation but lessens the amount of calcium (therefore there is some magnification by Ca-activated intracellular Ca release)
What is the mechanism of induction of NMDAR-dependent LTP? (two types)
Input-specificity, co-operativity + associativity
- LOW frequency stimulation: response primarily mediated by AMPARs (NMDARs has Mg block - inhibitory activity hyperpolarises the cell and intensifies Mg block)
- HIGH frequency transmission: sustained depolarisation - inhibition weakened by GABA-B autoreceptors - slowly activating NMDARs activated as Mg repelled, slowly contribute to depolarisation
How can input specificity be explain molecularly?
Active inputs required to release L-glutamate
*only get potentiation at synapses where glutamate binds NMDARs, adjacent inactive synapse will not be potentiated because not receiving glutamate
Referred to as Hebbian because Hebb postulated need for pre-and-post-synaptic activity; in former model referred to AP post-synaptically
* APs work to remove Mg block, but don’t actually need APs because process can work with just modest depolarisation * but Hebbian nature means needs to activate pre-synaptic to release glutamate, and activate post-synaptic to remove Mg block of NMDARs
How can cooperativity be explained molecularly?
Multiple inputs required for sufficient depolarisation of relieve Mg²⁺ block
*if activate NMDARs at single synapse, not enough NMDAR activation to induce plasticity (other synapses must be activated at the same time, for enough depolarisation to remove Mg block)
How can associativity be explained molecularly?
Depolarisation from other inputs
- Other types of non-glutamatergic synapse, may help to remove Mg block; neuromodulatory substance could help activate NMDARs
- many neuromodulatory NTs likely to have their effects by modulating NMDAR-dependent plasticity!
Broadly speaking, what is the role of of NMDARs in plasticity?
Their activation mediates most forms of LTP (NMDARs also contribute to synaptic transmission)
Also mediate lots of forms of LTD (different patterns of activation can lead to depression of synaptic transmission)
* info can be stored by decreasing or increasing synaptic weights
What are the typical induction parameters for LTP and LTD?
LTP:
1. Typically high frequency stimulation (HFS) / TETANUS (e.g. 100Hz for 1s)
- Theta bursts (priming) - few bursts at 5-10Hz
LTD: Prolonged low frequency stimulation (LFS) e.g. 1 Hz for 15 mins
