Saturday, May 18, 2013

Homeostatic platsicity in a thorny situation

Synapses, the connections between neurons can strengthen and weaken depending on the specific activity at that synapse. This is called synaptic plasticity, and we've talked about it a lot on this blog (here, here, here and here).

the strengthening and weakening of synaptic connections corresponds to the spine growing or shrinking (Matsuzaki 2007)

However, there is another kind of plasticity that can occur at synapses. This is called homeostatic plasticity. And instead of the synapse strengthening or weakening depending on the specific activity at that synapse, the synapses strengthen and weaken in homeostatic plasticity depending on the activity of the whole cell.

To drastically simplify, each cell 'wants' to fire about a certain amount, if it suddenly starts to fire a lot less, it will take steps to strengthen its connections or make itself more 'excitable' so it can get back to its preferred amount of firing. Similarly if the cell starts to fire a lot more than normal, it will take steps to make itself less excitable and to weaken its connections until it reaches the right amount of firing. 
Thorny Excrescences from Lee et al., (2013)
A recent paper from the Pak lab explains how in some specific neurons in the hippocampus (CA3 pyramidal cells), the activity of the whole cell is strongly controlled by a some very peculiar synapses. These synapses are close to the cell body, and are on these HUGE weirdly shaped spines (see above) called "Thorny Excrescences". For comparison 'normal' spines look more like this:
Spines from Lee et al. (2013)
The Thorny Excrescences (TEs) are massive spines that contain many separate synapses on them, but connect to the dendrite through 1 neck. 'Normal' spines, on the other hand, usually have 1 synapse at the spine head, and connect to the dendrite through 1 neck.

The size of the TEs, and their proximity to the soma makes them an extremely powerful way to control the signals that the soma receives. Lee et al (2013) shows that when you drastically reduce activity by blocking action potentials (using TTX), you get massive growth of these TEs, but the normal spines further away from the soma stay the same.

They test 3 things to determine whether the TEs have undergone homeostatic plasticity. They look at the morphology (they are bigger), the activity (the electrical signals from them are bigger) and the molecular signatures (the molecules indicative of new synapses are more plentiful). The paper is a really nice complete story showing that these TEs have a lot of control over the general activity of the cell.

It also solves an important problem with homeostatic plasticity. That is, how can the general activity of the cell be modulated without the specific differences between synapses being erased, and consequently the memories or pieces of information they encode? If homeostatic plasticity occurs at spines dedicated to it, then the other spines can still encode specific signals while the activity of the cell as a whole changes.

© TheCellularScale

Lee KJ, Queenan BN, Rozeboom AM, Bellmore R, Lim ST, Vicini S, & Pak DT (2013). Mossy fiber-CA3 synapses mediate homeostatic plasticity in mature hippocampal neurons. Neuron, 77 (1), 99-114 PMID: 23312519

1 comment:

  1. I've been doing some research (read: looking things up online) on nonsynaptic plasticity, of which homeostatic plasticity appears to be a part. (Correct me if I'm wrong about that.) I'm curious if there are any other known mechanisms by which neurons can alter their own intrinsic excitability.

    For example, is there some protein A that gets released in response to some stimulus X and works by, say, mucking about with the ion channels inside a neuron?