STDP Flashcards
• Levy and Steward (1983)
First documentation. Recording of left hippocampal slices. Receiving input from ipsi- and contralateral inputs from the entorhinal cortex. EC-DG pathway. Population EPSPs. Varied when activated the ipsi and contralateral
o LTP if concurrent or if contralateral first, LTD if ipsilateral precedes. 20ms window
o Just switching the order of plasticity of Glu synapses of the right pathway → timing crucial
• Stuart et al (1997
Patched neocortical layer 5, rat brain slices. Dual and triple patch recording. Dendrites, soma or cell body
o When AP generate, initially in AIS, it is capable of propagating forward as well as backward
o Backward AP is filtered into the dendrites (frequency dependant)
o TTX (v-gated Na+ channel blocker) → significantly reduces amplitude → active process. Not effected by cadmium.
• Magee and Johnston (1997):
Filled hippocampal CA1 pyramidal cell with Ca2+ sensitive dye. Patch recordings and calcium imaging.
o When induce AP in cell body → back propagate and can be seen as Ca2+ events in the dendrites
o Pairing subthreshold EPSP with back-propagating AP resulted in amplification of the EPSP and evoked Ca2+ influx
o If apply TTX over the dendrites → block propagation and LTP. When remove TTX then LTP could be produced
o Suggest that active signals are essential but not sure as to the mechanism
• Markram et al (1997):
probably the first study to describe STDP Dual whole cell voltage recordings pairing of cortical pyramidal cells, to control timing of APs in the 2 cells. (pairing post-synaptic APs and EPSPs)
o Pre-post (+ve delay) of 10ms → potentiation
o Post-pre (-ve delay) of 10ms→ depression. (using delays of 100ms, get little changes in EPSP) → relative timing important
o Blocked when add NMDA blocker
o Suggests that the result of synaptic integration, encoded within the back propagating AP. So single synaptic contact could be modified in the context of the summed activity of all the contacts of the neurons and temporal regulations.
• Zhang et al (1998):
In vivo whole cell recordings from developing Xenopus tadpole. Recorded from postsynaptic tectal neurons and presynaptic retinal cells
o Found same temporal relationship
o Synaptic inputs activated repetitively within 20ms before spiking of tectal neurons → potentiation. Subthreshold inputs activated within 20ms after spiking become depressed
o Suprathreshold inputs can therefore help to strengthen preceding or synchronously activated subthreshold potentials
• Froemke and Dan (2002)
also applies to the beyond STDP Studied way in which STDP operated in context of more natural spike trains. (Layer 2/3 rat) In vivo recordings of spike times and played them back in vitro visual cortical slice recording situations. Of 2 neurons in the cortex while animal is looking at visual stimuli→ what happens to the tuning preference of the cell
o Spike patterns seen in vivo can induce plasticity consistent with STDP
o Found very good correlation between predicted strength based on STDP algorithm and measured strength using standard protocol to duos
o Using complex patterns (triplets ) Direction and magnitude of synaptic modifications induced by spike patterns recorded in vivo in response to natural visual stimuli were well predicted by incorporating the suppressive inter spike interaction within each neuron→ not only dependant on the timing between neurons but also on the spiking pattern → for natural spike trains the timing of the first spike in each burst is dominant in synaptic modification.
o Suggested that this could be a physiological process of synaptic plasticity
o Dependence on pre/post interspike interval and suppressive interaction between consecutive spikes within each neuron provide a description of STDP
o - The spike trains of ‘natural’ were recorded from V1 neurons on anaesthetised cat not rat
o Suppression model: first spike of each neuron plays a dominant role in determining the sign and magnitude of synaptic modification E.G. for post-pre-post then LTD whereas Pre-post-Pre then LTP
• Yao and Dan (2001)
Visual cortex in humans. Neuron with one preferred orientation receiving inputs from neurons with different preferred orientations. Varied the interval between the conditioning pairs of stimuli
o Repetitive pairing of visual stimuli at 2 orientations induced a shift in orientation tuning of cat cortical V1 neurons, direction of shift was dependant on the temporal order of the pair. 40ms delay or less required
o Similar conditioning induced a shift in perceived orientation by human subject→ functional relevance. Relative timing of visual stimuli can play a critical role in dynamic modulation of adult cortical function
o Findings suggest can change orientation representation in the visual cortex and in perception. Significant shifts in orientation tuning
o Conditioning cat V1 neurons with preferred ORIETATION +15 preceding preferred orientation induced shift toward whereas conditioning of S-15 and S also did the same. Conditioning in reverse order caused tuning to shift away from the non optimal orientation
o Humans sinusoidal gratings were used, looking at perceived orientation
o Clearly see the perceptual shift they were able to produce was only evident when→ delivering switch in timescale that was relevant for STDP
o Supports STDP in humans
• Navian and Sakmann (2004):
Patched pyramidal cells and filled with Ca+ dye. Layer 4 of somatosensory barrel cortex of young rats. Thalamocortical brain slices. 2 photon microscopy
o Pre-post → supralinear response (greater than would be predicted as linear summation of both pre and post Ca2+ events)→ consistent with idea that satisfy the coincidence detection for NMDA and get large influx of Ca2+.
o - Not actually looked at NMDA receptors (e.g. if use AP5 does it block the tLTP induction) (did find that use of NBQX to block AMPAR the response maintained)
• Rodriguez-Moreno and Paulson (2008)
Layer 4 and L2/3 neurons in mouse barrel cortex slices, Loading MK-801 into pre or postsynaptic neurons, paired recordings. L4 stimulation and EPSPs of L2/3 pyramidal cells Used NMDA receptor blocker MK-801 in the post synaptic recording pipette and in the presynaptic recording pipette
o Synaptic potentiation requires post synaptic but not presynaptic, NMDA receptors. Whereas synaptic depression requires presynaptic but not post-synaptic receptors
o AP5 blocked both tLTP and tLTD
• Froemke et al (2005):
(also good for location) Rat cortical layer 2/3 pyramidal neurons, brain slices . Whole cell recordings. Stimulated presynaptic spike trains
o Magnitude and temporal specificity of STDP vary along the dendrite. Distal dendrite the magnitude of LTP is small and the window for LTD Is broader
o 2 photon imaging revealed that Ca2+ signals were larger distally than proximally (suggests larger Ca2+ signals may cause longer suppression and widen LTD windows)
o Spike timing window for LTD correlates with he window of AP induced suppression of NMDAR (correlation applies to both their dendritic location dependence and pharmacological properties )
o Found significant suppression of EPSPs in AP→EPSP pairs → sublinear summation of EPSPs, compared to supralinear for pre-post
o Presynaptic stimulation with partial blockade of NMDA receptors induced LTD and occluded further induction of STDP LTD→ suggests that NMDA receptor suppression underlies LTD induction.
o Blocked with AP5
o Location dependant tuning of input could enhance the computational capacity of cortical pyramidal neurons
o Loaded postsynaptic neuron with BAPTA, Ca2+ chelator, and measured AP induced suppression of NMDA-EPSPs → blocked the suppression indicating that NMDAR suppression is Ca2+ dependent → The suppression is also largely abolished by bath application of the L-type calcium channel antagonist nimodipine
• Rosenmund et al (1995):
Whether synaptically activated NMDARs are regulated by intracellular Ca2+ in cultured hippocampal neurons by comparing EPSPs to the previously described Ca2+ dependant regulation of whole NMDA currents. Whole cell recordings
o Calcium entry during NMDAR mediated EPSP was sufficient to produce reversible Ca2+ dependant inactivation. Suggest that Ca2+ dependant regulation of synaptic NMDAR can control the amplitude and duration of EPSC and thus regulate Ca2+ influx on an impulse-to-impulse basis
o Study suggest that NMDARs are inactivated by intracellular Ca2+ and Ca2+ entry through synaptically activated NMDARs is sufficient to provide feedback inhibition of slow EPSP
o Decrease in slow EPSP were reduced in low Ca2+ extracellular solution
o The decrease in EPSP was dependant on the extracellular Ca2+ concentrations→ consistent with the desensitisation/inactivation of postsynaptic NMDARs
o NMDA channel activity was measured by whole cell application of NMDA before and after synaptic stimulation→ a series current was reduced
o - doesn’t directly show that NMDAR are inactivated
Woodin et al (2003)
): GABAergic synapses in hippocampal cultures and slices. Measuring GPSCs
o Repetitive post-synaptic firing within 20ms before and after the activation of GABA synapses lead →LTP
o Synaptic modification required Ca2+ influx through post-synaptic LTCC, (nimodipine) and was due to local decrease in K/Cl cotransport activity (furosemide) , effectively reducing the strength of inhibition
o At longer intervals such as ~100ms get depression.
• D’amour and Froemke (2015
Layer 5 cortical neurons of rat auditory cortex slices. Study how Glu and GABA synapses on same post-synaptic neuron may be being regulated by the interplay of different types of STDP. Separate out to measure strength of excitatory (Picrotoxin to block GABA) and inhibitory synapse (DNQX to block AMPAR) input onto the same cell. Whole cell recordings
o Pairing pre and postsynaptic spikes → potentiated inhibitory inputs (~10ms). Excitatory synapses display a asymmetrical STDP time widow
o Both required NMDAR activation and adjusted the excitability-inhibitory ratio of events paired with post-synaptic spiking
o Time window between excitation and inhibition became more precise
o Argue that interplay between different STDP function allow a balance between excitatory and inhibition onto common post-synaptic neuron
o So if pathway has lower E/I synaptic plasticity tend to balance so the and vice versa→ so get matching between excitation and inhibition for a particular pathway
o We found that spike pairing shortened the integration window between excitation and inhibition regardless of the temporal order of pre- and postsynaptic spiking. This suggests that inhibitory STDP enforces spike-timing fidelity by reducing the period during which incoming events are effective in depolarizing postsynaptic neurons. Additionally, STDP increased the reliability of paired inputs evoking action potentials after pre/post pairing, but decreased spike firing probability after post/pre pairing.
o - How does NMDAR signalling interact with postsynaptic for induction of excitatory and inhibitory long term plasticity
o -How are synapses sensitive to the specific E/I ratio during pairing ?
• Sjostrom and Hausser (2006):
Layer 2/3 and L5 inputs to pyramidal L5 neurons. Looked at whether enhancing back propagation from distal sites of the soma, could change the LTD to LTP. Added a source of dendritic depolarisation (electrical stimulus) what was delivered in coincidence with bAP Patch clamp recordings, 2 photon imaging
o Tendency of synapses close to the body to be potentiated and synapses further away to be depressed.
o Augmenting bAP led to potentiation→ show that STDP function can vary depending on location of the synapse relative to the soma.
o 3rd factors can modulate these process→ in this case other source of depolarisation resulted in LTP not LTD
o Sign of synaptic is regulated by the spread of the bAP to the synapse → creates a progressive gradient between LTP and LTD as the distance from the soma increases
o Electrical interactions are reflected in terms of dendritic Ca2+ signals – demonstrate that the switch is linked to highly supralinear increase in dendritic Ca2+
• Zhang et al (2009):
Glutamatergic synapses of the hippocampus. Explored STDP function with and without applying DA to cortical slices. 20μm DA
o Pairing STDP protocol of +45ms that would not normally produce STDP. When apply DA then get potentiation → DA makes the timing window longer
o Negative delay -10ms (should lead to LTD) actually led to LTP
o DA seems to favour LTP
o Allowed normally ineffective weak stimuli with fewer spike pairs to induce tLTP
o Was completely blocked in with D1-like receptor antagonist but not D2-like receptor antagonist. Blocked by NMDAR antagonist
o Indicates that D1 like receptor mediated modulation appears to act through the classical NMDAR-mediated signalling pathway,→Consistent with global reward signal
