There are many forms of plasticity that operate in the brain. The term plasticity refers to the ability of neurons to change their response to a signal based on what has happened before. Such a change in response can last for a short time (on the order of miliseconds to seconds) or for a long time (on the order of minutes, hours, days ..... years even!!). How do we know? Well, we can record the electrical activity of individual neurons. The currents are very small and it requires very specialized equipment to do it, but many experiments over many years have shown that many neurons in many parts of the brain undergo changes in the size or frequency of response to an incoming signal. Two of the most studied form of plasticity are long-term potentiation (LTP) and long-term depression (LTD), which are seen as an increase and decrease in the size of the response, respectively.

LTP and LTD - memory mechanisms?

LTP and LTD have been considered to be possible mechanisms underlying learning and memory. Both LTP and LTD can be induced by a single round of synaptic activation and the change in synaptic weight is maintained over a long period of time. These are both properties one would expect to see in a molecular memory mechanism. So, is there any evidence that this may be the case?

Reprsentation of Clathrin-mediated internalisation
Representation of clathrin-mediated internalisation of AMPA receptors. AP-2 is represented by the blue discs while clathrin is represented by the yellow buckyballs. © University of Bristol
It is extremely difficult to link a specific process occurring in a neuron with a specific memory. You would have to be able to record the activity of that neuron (or group of neurons) while an individual animal was learning or remembering something ... and how do you know which one to record from ... or from which bit of the brain? Hence an approach is required that pieces together 'circumstantial' evidence that links a molecular/cellular mechanism with a memory process - and the best evidence to date is work carried out here at the MRC Centre linking visual recognition memory with LTD.

LTD and Visual Recognition Memory

Visual recognition memory has long been associated with a lowering of neuronal activity. When an object is seen for the first time, some neurons in the perirhinal cortex have a strong response which is much lower when the object is seen again. This 'decremental' response signals that the object is now familiar and so is referred to as familiarity discrimination. Similarly, an animal will normally investigate a new object for longer than one that is familiar to it - the spontaneous object recognition task. This response can be blocked by interfering with the molecular process that underlies all forms of LTD - clathrin-mediated internalisation of AMPA receptors. This process is a universal mechanism for the removal of proteins from the cell surface where a cage of clathrin protein is built around an invagination of the cell membrane. This is initiated by another multi-subunit protein called AP-2. It interacts with the GluA2 subunit of the AMPA receptor at a site where a stabilising protein called NSF also binds. AP-2 replaces NSF, resulting in receptor internalisation.

Animation showing how a competing peptide against AP-2 can block internalisation. © University of Bristol
The interaction between AP-2 and the GluA2 subunit can be blocked by using a small part of the GluA2 protein that includes the binding site, a peptide. If a high concentration of such a peptide is flooded into a cell, it binds to all the AP-2 proteins, preventing it replacing NSF and hence blocking receptor internalisation. We have shown that if this is done in individual neurons in brain slices, LTD is blocked. We have then been able to show that animals that express a competing peptide in the perirhinal cortex cannot distinguish familiar from novel objects in the spontaneous object recognition task and that LTD is absent in perirhinal cortical neurons from these same animals. Thus we have one change (presence of AP-2 blocking peptide) that blocks both LTD and familiarity discrimination. This is the best evidence to date linking a specific molecular mechanism with a specific memory process.