Compete off-beat to delete!

Filed under the category of “interesting things neurons do...”

A forthcoming paper by Favero and colleagues showed how motor neurons compete with each other for synaptic “real estate.” The fastest, most effective way for a “winner” to be decided was not to have the two neurons firing “head to head” at the same time, as the figure below shows (a composite of parts of the their Figures 1 and 2):

First, let me explain what “polyneuronal innervation” on the left Y axis means. Favero and company were working with rats. Normally, muscle cells in mammals (including rats) have just one motor neuron connected to them and controlling contraction in that muscle cell. But you can create a situation where more than one neuron innervates a muscle cell temporarily. That’s polyneuronal innervation, given in the left label.

The muscle here is the rat soleus muscle, which is innervated by two nerves. The team crushed the two nerves leading to the muscle, and the nerves grow back to rejoin the muscle. When they do so, the neurons don’t neatly divide up the muscle cells immediately. Many of the muscle cells get neurons from both nerves innervating them.  You can see this polyneuronal innervation in this picture (their Figure 5):

The key things are the three “blobs” that show a mix of red and green. Those are the synapses between the muscle and the neurons. Neurons from one nerve are stained red, and those from the other are stained green.

This is not normal for adult mammals, and so you want to get rid of the polyneuronal innervation.

The researchers were able to control when the nerves innervating generated action potentials through a series of complicated interventions.

When the nerves were fired 20 ms apart or more (black bar at right in the top picture), you got much less polyneuronal innervation than when the nerves were fired at the same time (red bar at left).

Favero and colleagues do a series of experiments to show that the key factor here is the timing of the two signals, not just the muscle activation. They also showed that the relative number of spikes is also playing a role in determining how quickly you can get rid of those polyneuronal connections.

How general this phenomenon might be depend on how important competition between neurons is for determining synaptic connections. This polyneuronal innervation is not normal for adults, but the authors mention you do see it during development, so this is not a completely artificial scenario. Some authors, notably Gerald Edelman, have proposed that neuronal competition is a key element in shaping vertebrate brain circuits. This finding may have broad implications, but we’ll need some more good systems showing neural competition to test it.

Reference

Favero M, Busetto G, Cangiano A. 2012. Spike timing plays a key role in synapse elimination at the neuromuscular junction. Proceedings of the National Academy of Sciences: in press. DOI: 10.1073/pnas.1201147109

Picture from here.