LTP

Question 1

Bliss and Lømo

Answer 1

1973, discovered that short bursts of high frequency stimulation in the DG of rabbit slice generated long-lasting augmentation in the synaptic response between pairs of cells.
This was rapidly induced, and lasted for at least as long as the experiment was carried out (16+ weeks)

Question 2

Andersen

Answer 2

1980, found that LTP was input specific, such that only the tetanised pathways was enhanced. Thus, can change synaptic performance at any one synapse independently of all the other synapses making contact with the neuron.
This gives a great amound of computation power/freedom to the neuron

Question 3

McNaughton

Answer 3

1978, found that LTP may require simultaneous activity at neughbouring synapses, as one synapse alone may be insufficient to give rise to adequate depolarisation

Question 4

Barnioneuvo

Answer 4

1983, evidence for associativity (similar to cooperativity), such that concurrent activation of a weak and strong input can potentiate the weak input under conditions where stimulation of either input alone did not

Question 5

Collingridge

Answer 5

1983, Evidence that LTP is NMDAR-dependent in this pathway
− Stimulating schaffer collateral-commissural projection and recording in stratum radiatum in the CA1 region of rat Hc slices
− APV (AP5, NMDAR antagonist) prevented LTP evoked by high frequency stimulation, but had no effect on basal performance of the synapse
− Once drug was washed away, LTP could be induced once more
− Thus, although NMDARs do not appear to be involved in normal synaptic transmission in this pathway, they may play a role in synaptic plasticity
Another glutamate receptor must be involved in general excitatory transmission

Question 6

Nowak

Answer 6

1984, Evidence that NMDARs are Mg gated, and that they act as coincidence detectors
− Patch-clamp recordings of mouse embryo central neurons in response to glutamate
− Under resting and basal conditions, the receptor remained closed even if glutamate were bound to it glutamate not sufficient to open receptor channel
− If receptor was depolarised, receptor would open when glutamate was bound
− Found a link between voltage sensitivity and Mg2+ sensitivity
− In Mg2+-free solutions, agonists open the voltage-dependent NMDARs
− In presence of Mg2+, the single-channel currents measured at resting potential are chopped in bursts and the probability of opening the channels is reduced
− The voltage dependence of the NMDAR-linked conductance appears a consequence of the voltage dependence of the Mg2+ block of the NMDAR
− So, NMDARs are coincidence detectors

Question 7

Malenka

Answer 7

1988, evidence that Ca is both necessary and sufficient for LTP.
− As Ca2+ flux is unique to NMDARs and NMDARs are required for LTP induction, it was thought that Ca2+ flux may play a role in LTP induction
− Critical role of postsynaptic calcium in triggering LTP examined in three experiments
− Nitr-5 (a photolabile calcium chelator, releases calcium in response to UV light) injected into hippocampal CA1 pyramidal cells. Photolysis (uncaging, thus liberating high levels of intracellular calcium) resulted in a large enhancement of synaptic transmission
− Buffering intracellular calcium at low concentrations blocked LTP
− Depolarisation of the postsynaptic membrane so that calcium entry is suppressed blocked LTP
− thus, increase in postsynaptic calcium is necessary to induce LTP, and sufficient to potentiate synaptic transmission

Question 8

Manilow (1)

Answer 8

1989, evidence that CaMKII is necessary for LTP induction
− LTP induction is blocked by injection of the following into postsynaptic CA1 pyramidal cells:
− H-7 (general protein kinase inhibitor)
− CaMKII(273-302) (selective CaMKII inhibitor)
− Once established, LTP is unresponsive to postsynaptic H-7
− So, postsynaptic PKC and CaMKII are required for induction of LTP, but not for the maintenance of LTP

Question 9

Silva

Answer 9

1992 (a+b), evidence that CaMKII is necessary for LTP induction and for spatial learning
− Using a KO animal for the first time in the field of neurobiology- so can do behavioural experiments
− KO mice that don’t express alpha-CaMKII (highly enriched in postsynaptic densities of hippocampus and neocortex)
− KO mice exhibited mostly normal behaviours, presented no obvious neuroanatomical defects. Whole cell recordings show postsynaptic mechanisms (including NMDAR function) are intact
− However, KO mice are deficient in ability to produce LTP
− KO mice exhibit specific learning impairments, indicating alpha-CaMKII has prominent role in spatial learning, but is not essential for some types of non-spatial learning

Question 10

Lee (1)

Answer 10

2003, evidence that CaMKII activation is spine-specific
− Monitored spatiotemporal dynamics of CaMKII activation in individual dendritic spines during LTP using 2-photon fluorescence microscopy and 2-photon glutamate uncaging
− Found that LTP induction triggered transient (1min) CaMKII activation restricted to the stimulated spines
− CaMKII was specifically activated by NMDAR and L-type VGCC in response to Glu uncaging and depolarisation

Question 11

Shen and Meyer

Answer 11

1999, used GFP-tagged CaMKII to observe the translocation from the cytosolic F-actin-bound state to the PSD-bound state
− Autophosphorylation of CaMKII indirectly prolongs its PSD localisation by increasing the calmodulin binding affinity

Question 12

Opazo

Answer 12

2010, evidence that CaMKII triggers the trapping of surface AMPARs through phosphorylation of stargazin
− Used quantum dot labeling to track AMPAR subunit positions
− Show that CaMKII activation and postsynaptic translocation induce the synaptic trapping of AMPARs diffusing in the membrane
− AMPAR immobilization requires phosphorylation of stargazin and its binding to PDZ domain scaffolds
− Immobilization is not seen when using non-phosphorylatable stargazin or when using a CaMKII variant which is unable to bind NMDAR
Showed that immobilization doesn’t depend on the phosphorylation of GluR1 at Ser831

Question 13

Ashby

Answer 13

2006, evidene that lateral diffusion of AMPARs has a role in AMPAR trafficking
− Selectively visualized surface-expressed AMPAR by tagging GluR2 subunits with pH-sensitive GFP (SEP)
− Using fluorescence recovery after photobleaching (FRAP)
− Showed that lateral diffusion is responsible for the continual exchange of a substantial pool of AMPARs at the spine surface
− Showed that lateral diffusion depends on spine morphology and is limited by the spine neck (protein movement is slower in/out of spines compared to non-spines)
− Thus, evidence for the role of lateral diffusion in AMPAR trafficking, and gives explanation for how spine structure can maintain synapse-specificity of signaling

Question 14

Malenka and Südhof

Answer 14

2017, found evidence for a role of Syt1 and Syt7 in AMPAR exocytosis during LTP
− In mouse Hc PCs, blocked postsynaptic expression of both synaptotagmin-1 and 7 (Syt1, Syt7)
− This did not impair basal synaptic transmission or alter AMPAR trafficking events
− But, did abolish LTP expression
− Could be restored by expression of WT Syt7, but not of Ca-binding deficient mutant
− They suggest that postsynaptic Syt1 and Syt7 act as Ca-sensors for Ca-dependent AMPAR exocytosis during LTP

Question 15

Derkach

Answer 15

1999, showed that phosphorylation of Ser-831 in GluR1 by CaMKII potentiates the receptor current

Question 16

Lee (2)

Answer 16

2003, foudnt hat mice with knockin mutations in the GluR1 phosphorylation sites show deficits in LTP (+LTD) and spatial learning tasks, indicating that GluR1 phosphorylation is necessary for LTP expression

Question 17

Granger

Answer 17

2013, argued that Lee's LTP effect is independent of subunit type, and that LTP requires a pool of Glu receptors
−	Used single-cell molecular replacement strategy in mouse CA1 Hc pyramidal neurons to replace all endogenous AMPAR with transfected subunits
−	Found no requirement for GluA1- replacement with GluA2 showed normal LTP
−	Indeed, GluK1 is sufficient for mediating LTP (KARs are a separate class of glutamate receptors that differ in fundamental ways from AMPARs). Showed this by substituting KARs for all AMPARs in a KO experiment

Question 18

Manilow (Granger response)

Answer 18

argued that genetic deletion of GluR1 is not physiological and must fundamentally change synapse activity, negating Granger’s findings.
Also, just because Kainate receptors can be inserted, this does not mean that this occurs under physiological conditions (may be a compensatory mechanism)

Question 19

Zakharenko

Answer 19

2001, found that FM1-43 unloading is accelerated after LTP induction, giving evidence for presynaptic changes
− Imaging changes in presynaptic function at boutons of CA3-CA1 exctitatory synapses in acute hippocampal slices
− (FM dyes = used to image synaptic recycling. Not fluorescent in aqueous solution, but fluoresce when associated with the inner leaflet of synaptic vesicles (lipid). This fluorescence can then only be lost by exocytosis)
− Showed enhanced presynaptic function during LTP induced chemically or by electrical stimulation
− Thus, LTP is likely to depend, at least in part, on enhanced transmitter release from the presynaptic neuron

Question 20

Bayazitov

Answer 20

2007, also found evidence for a presynaptic component, though with a less flawed technique
− Used a transgenic mouse that expresses spH in CA3 PCs
− (spH = fusion protein of VAMP2/synoptobrevin and pHlourin (pH-sensitive GFP). Increases it’s fluorescence when synaptic lumen is exposed to extracellular space. It integral part of vesicle’s membrane, so is not release (can be used multiple times!))
− monitored presynaptic vesicle exocytosis whilst monitoring postsynaptic potentials
− found that CA3-CA1 LTP induced by 200Hz tetanisation or theta-burst stimulation consisted of both slow presynaptic and fast postsynaptic components

Question 21

Schuman and Madison

Answer 21

1999, argued that NO may act as a retrograde messenger for presynaptic LTP induction
− Extracellular application or postsynaptic injection of nitric oxide synthase inhibitors blocks LTP
− Extracellular application of haemoglobin (binds NO) also attenuates LTP
− Findings suggest that NO liberated from postsynaptic neurons may travel back to presynaptic terminal to cause LTP expression

Question 22

Manilow (2)

Answer 22

1989, showed that CaMKII is not necessary for the maintenace of LTP as postsynaptic H-7 did not block LTP maintenance (though did block induction)

Question 23

Nguyen

Answer 23

1994, found evidence for the requirement of a critical period of transcription for l-LTP induction
− Used different inhibitors to show that in rat hippocampal slices, the induction of L-LTP (by tetanic stimulation of application of cAMP analog) was selectively prevented when transcription was blocked immediately after tetanisation/cAMP analog application
− Transcription requirement had critical time window

Question 24

Ling

Answer 24

2005, found that PKM(zeta) is both necessary and sufficient of LTP maintenance
− Postsynaptic inhibition of typical PKC by staurosporine does not effect LTP maintenance (but does effect induction)
− Pharmacological inhibition of atypical PKMzeta by chelerythrine inhibits the maintenance of established LTP
− Zeta inhibitory proteins (ZIPs) reverse established LTP invitro and in vivo
− Diffusion of (active) PKMzeta into cells enhanced ESPC amplitudes within 6minutes (heat inactivated PKMzeta had no effect)
− A non-NMDA glutamate receptor antagonist suppressed PKMzeta-augmented EPSCs, indicating PKMzeta acts to enhance AMPAR currents

Question 25

Volk

Answer 25

2013, using a genetic method, show that PKM(zeta) is not necessary for Hc LTP, learning and memory
− Generated transgenic mice lacking PKCzeta and PKMzeta (mice were both viable and fertile, and showed no anatomical abnormalities)
− Created both conventional and conditional KO mice to limit possible compensatory effects during development
− These KO mice show normal synaptic transmission and LTP at schaffer collateral-CA1 synapses
− Furthermore, KO mice have no deficits in several Hippocampal-dependent learning and memory tasks
− Furthermore, ZIP still reverses LTP (inhibits maintenance) in these KO mice, indicating that the effects of ZIP are independent of PKMzeta

Question 26

Saktor lab (Ling)

Huganir lab (Volk)

Answer 26

In 2016, Saktor lab argued that in the case of PKM(zeta) KO, PKM(lambda), which is not normally involved in palsticity, is replacing it

Huganir lab now looking to create double KO mouse

Question 27

Frey and Morris

Answer 27

1997, first evidence for the synaptic tagging and capture hypothesis
− propose that LTP initiates the creation of short-lasting protein-synthesis-independent ‘synaptic tags’ at the potentiated synapse, which sequesters the relevant proteins travelling from the soma to establish late LTP
− two stimulating electrodes placed in independent pathways innervating the same population of rodent CA1
− repeated strong stimulation of S1 would lead to L-LTP
− weak tetanic stimulation, or repeated stimulation in the presence of protein-synthesis inhibitors, of S2 lead only to E-LTP
− But, such stimulation of S2 could result in L-LTP so long as S1 L-LTP had been generated within previous 3hrs
− Indicates that generation of L-LTP depends not only on local events during induction, but also on prior activity of the neuron
− In addition, later studies showed that E-LTP could be converted to L-LTP if S2 preceded S1 by up to an hour

Question 28

Hayashi and Majewska

Answer 28

2005, found evidence that spine geometry changes with LTP
− The addition of receptors at the synapses causes the synapses to grow in size (in order to accommodate addition of AMPARs)
− When induce LTP at dendritic spines (do it one dendritic spine at a time), you observe dendritic growth at that one spine
− (they speculate that they may shrink back in size with LTD, though the evidence is less good for this)
− evidence that the geometry of dendritic spines controls postsynaptic calcium signalling and is bidirectionally regulated during synaptic plasticity

Question 29

Myer

Answer 29

2014, found evidence for two stages of structural changes
− Induced spine enlargement by 2-photon glutamate uncaging
− Examined the relationship between spine, PSD and bouton size by 2-photon time-lapse imaging and electron microscopy
− Found that synaptic activity first leads to an immediate increase in spine volume, which is matched by an increase in Homer1c (over 30mins)
− Then, over the next 1-3hrs, the spine either retracts to original size (no persistent enlargement), or PSD-95 also accumulates and both the PSD and presynaptic bouton increase in size as well, leading to stably enlarged synapses
− Thus, they argue that there is an early phase of structural LTP associated with fast spine growth, and a later phase associated with growth of PSD and bouton

Question 30

Padamsey

Answer 30

2016, found evidence that the exocytosis of Capthesin B regulates structural LTP of spines
− Found that bAPs could trigger Ca release from lysosomes
− This Ca release triggered the fusion of lysosomes with the palsma membrane, causing release of Cathepsin B
− Capthesin B increased the activity of matrix metalloproteinase 9 (MMP-9), which is involved in ECM remodeling
− Inhibition of lysosomal Ca signaling or Capthesin B release prevented maintenance of dendritic spine growth, but this could be rescued by application of active MMP-9
− Indicates Capthesin B release and MMP-9 activation has a role in regulating sLTP of dendritic spines