week 5 - CAMK2 autophosphorlyation Flashcards
What was the original CAMK2 hypothesis?
At a synapse a kinase might phosphorylate itself so it becomes autophosphorylated.
This might change the kinase into a persistently active activity mode and this persistent activity might maintain memory.
This came about because molecules are turned over constantly, so for a memory to be maintained the need to be a mechanism that can maintain memory despite molecular turnover
CAMK2 has subunits which each has a catalytic domain to phsophroylate subunits, and each subunit can be swapped out, but the autophosphorylation still takes place allowing the molecule to maintain persistent activity.
Originally, CAMK2 was proposed to be a molecular mechanism to store memories in the brain
What is CAMK2? Talk about how it works
CAMK2 is a major post-synaptic density protein.
It has a catalytic domain and regulatory domain
When Calcium levels rise, Calcium Calmodulin binds to the regulatory domain.
This leads to a structural change so Camk2 can phosphorylate substrates. This is dependent on calcium, so if calcium levels drop, CAMK2 is inactive again
However, CAMK2 can also phosphorylate itself at a specific site (T286). Once phosphorylated here, the kinase remains active even if calcium levels drop
What is the best way to test the hypothesis that CAMK2 autophsophorylation is a molecular mechanism of memory storage
Loss of function:
Block the autophosphorylation of CAMK2 at T286
Gain of function:
If we stimulate autophosphorylation of CAMK2, can we generate an artifical memory
What did malinow et al show about CAMK2
Malinow et al. showed that blocking postsynaptic CaMKII (calcium/calmodulin-dependent protein kinase II) function prevents the induction of long-term potentiation (LTP).
CAMK2 was blocked by using a peptide that mimics the inhibitory site of CAMK2 to block CAMK2 activity
When CAMK2 was blocked, there was only a transient synaptic potentiation lasting about 30 minutes and then there was no potentiation any longer
This is consistent with the idea that CAMK2 is important for LTP
CON:
This peptide also blocked protein Kinase C, so unfortunatly it wasn’t specific meaning this study couldn’t be conclusive
What are genetic approaches to study LTP
Pharmacological drugs are only available for a few proteins. Therefore,
pharmacology cannot address the role of many molecular processes
important for LTP and study their role in memory.
However, molecular genetics can result in manipulation of any protein in
particular ways: 1) genes can be knocked out, 2) knockouts can be restricted
to cell types and/or brain regions, 3) knockout can be restricted in time, 4)
point mutations can be introduced, 5) proteins can be overexpressed.
Why is it tricky knocking out CAMK2 in mice
Because they have different genes/isoforms
Most knockout mice just knockout the Alpha subunit
Evidence that CAMK2 is important for spatial memory
alphaCAMK2 knockout mice showed impaired performance in the morris water maze, suggesting that ACamk2 is nessercary for spatial memory formation
How did Elgersma (2002) update the CAMK2 knockout mice
What did Elgersma et al. (2002) discover about inhibitory autophosphorylation of CaMKII?
They created T305D mutant mice where they prevented Caclium Calmodulin from binding to CAMK2
They compared them with KO mice. They found that there is no LTP in the T305D mice, but there is still some residual LTP in the KO mice
This Experiment CLEARLY showed that CAMK2 is critical for LTP
This also translated to behaviour:
KO mice showed mild learning/memory deficit
T305D mice showed Severe Deficit
How was it determined that that aCAMK2 knockout deficits is due to memory NOT developmental abnormalitys
Atchbernbag et al waited until the animal was fully developed normally, then managed to inject tamoxiphine to knockout the gene after development.
Tested animals in the water maze and found that the KO animals had no memory for the spatial platform location
This shows that the memory deficits in the original knockout was not due to developmental abnormalities
what is the role of CAMK2 autophosphylation?
Mayford et al (1996)
Mice express a transgene as well as endogenous gene. T286D mutation mimics autophosphorylation of alphaCaMK2
These mice also have impaired spatial learning
How to CAMK2 trangene expression levels link to LTP (Bejar et al, 2002)
When transgene expression levels are low, LTP is impaired
read more on this paper
What did mayford et al 1996 discover about the role of CAMK2 autophosphorylation?
They made a transgenic mouse that have a gene mutation that mimics the camk2 autophosphorylation at T286, ONLY when doxycycline is present
These mice also show impaired LTP and spatial learning when doxycycline is present
This was a suprise, because one would have assumed that a gain of function would occur, according to the CAMK2 autophosphorylation hypothesis
This effect may be due to the effect that such high levels of the transgene mean the synapse beccome saturated so you can’t produce more LTP
This was confirmed by Bahar et al, 2002, how found that whilst when transgene levels were high LTP was impaired, when it is low, the LTP was enhanced as predicted
What happens with alpha CamK2 point mutant mice (giese et al, 1998)
EVIDENCE CAMK2 autophosphorylation is REQUIRED for memory and LTP
alphaCaMK2 point mutant mice have SEVERLY impaired CA1 LTP and spatial learning and memory formation
The point mutation blocks T286 in the endogenous gene, meaning CAMK2 cannot be autophosphorylated
Shows the autophosphorylation of CAMK2 is ESSENTIAL for memory and spatial learning
however, Cooke et al (2006) found that whilst this was the case in CA1, it was not in the case in the dentate gyrus in aneastetised mice
HOWEVER
They think the reason for this is that there is neurogeneis in the dentae gyrus, and not in CA1. The new neurons in the dentae gyrus make new initial LTP signalling pathways that do not require CAMK2
MAIN CON OF THE WORK:
- They have not used inducible techniques for gain of function
- This means its unclear where autophosphorylaition is needed for MAINTENENCE OF LTP, or memory STORAGE. This is because for maintenence youd need to block autophosphorylation AFTER the memory is created
Is CAMK2 autophosphorylation required for memory STORAGE?
- Recently, a peptide inhibitor that blocks the activity of CAMK2 was invented
- This allowed researchers to test whether CAMK2 autophosphorylation is needed for memory STORAGE, because it meant that you can administer the peptide inhibitor AFTER the memory was created
- Buard et al foound that administering the peptide pre-training led to less freezing, but administering it AFTER ldid not
- This suggests that CAMK2 autophosphorylation is nessercary for memory INDUCTION but not memory STORAGE
How do we know that CAMK2 autophosphorylation is needed for ltp INDUCTION
When Giese et al tested mice 10 seconds after high frequency tetanus, the mice that had alphaCaMK2 point mutation, which blocks T286 in the endogenous gene, meaning CAMK2 cannot be autophosphorylated, had a lower fEPSP
This suggests the autophosphorylation of CAMK2 is REQUIRED for the INDUCTION of NMDA dependent LTP
OBSERVATIONAL EVIDENCE:
- Is CAMK2 phosphorylation increased for a long time after LTP induction?
MIXED RESULTS:
Barria et al, 1997 did find an increase in CAK2 photophrylation in synapses that had undergone LTP
Fukunaga et al found that activity increased following autophosphorylation, and this persisted for at least 60 minutes
HOWEVER
Lengyel et al found that after high frequency stimulation, the Activity of CAMK2 decreased
what did yasuda et al (2003)
They designed a flourescent reporter that records the activity of CAMK2
They looked at the activity of CAMK2 during LTP induction.
They found that CAMK2 activity initially increases however this declines within 60 secoonds
This is evidence AGAINST the CAMK2 autophosphorylation hypothesis as a mechanism for LTP
How important is CaMK2 autohphophorylation for learning and memory? (Need and Giese, 2003)
Very important. There is no spatial memory formation in T286A mutants
What is the conclusion from these studies?
this study shows that autophosphoryation is essential for induction of NMDA receptor-dependent potentiation in area CA1 and spatial memory formation
What is evidence against the CAMK22 hypothesis for memory storage?
Baurd et al, 2010)
CN21 is a peptide inhibitor which blocks camk2
When introduced before memory formation, it blocked learning, suggesting that camk2 is nessercary for formation
When introduced after formation, it did not block mainentence suggesting it is not nessercary for memory maintenence
Is LTP associated with persistent, autonomous CAMK2 activity? (fukunaga et al, 1993, chang et al 2017)
experiments are in conflict with each other.
Chang (2017) used imaging to look at this.
Found that in the dendritic spine, CAMK2 activity spikes but then decreases quickly after 20 seconds
what does this mean? watch the lecture again
Tullis et al 2023 - interaction camk2 and NMDA
The INTERACTION between CAMK2 and the NMDA receptor is fundamentally important for the induction of LTP
The autophosphorylation of CAMK2 is nesasercary for it to INTERACT with NMDA
They used an inhibitor that blocks kinase activity
They used another inhibitor that blocked the interaction between CAMK2 and NDMA
they found:
- Blocking the KINASE activity alone DID NOT impair LTP
- Blocking the association between CAMK2 and NDMMA blocked LTP
What is evidence for the CAMK2 hypothesis of memory storage (rossetti et al 2017)
Memory erasure hypothesis
By using a dominant negative blocker of CAMK2, k42M mutation makes CAMK2 inactive
METHODS
-Trained animals on a task where Animals get rotated on a platform and learn not to get rotated into the danger zone due to footshock
- After training, animals get injected a virus into the hippocampus that inhibits CAMK2
- after use of the CAMK2 blocker, the memory was erased
CONCLUSION
- CAMK2 IS memory storing
- Because CAMK2 was blocked AFTER the memory was created and the memory was ERASED
PROPOSED EXPLANATION
- After LTP induction, you get autophosphorylation of CAMK2
- Autophosphorylated CAMK2 associates with the NMDA receptor
- This association provides strctural information to maintain LTP and memory. What this is is speculative, but it might be that it keeps AMPA receptors at the synapse
