The Starburst Neuron is Complete!

We’ve finished our first starburst amacrine cell!

This was a very challenging neuron, because its dendrites were so thin and difficult to trace. Even after we had mapped most of the neuron, we kept on discovering new branches every now and then.  That’s why we kept mapping for so long. It’s time to map connections.

Now that the cell is complete, we can admire its beauty, and be proud of our accomplishment! The starburst neuron is thin and flat, like a pancake. Already you can tell that it’s different from the neurons you might have seen in textbooks.

SAC pancake, EyeWire, neuron, 3D

A textbook neuron has two kinds of branches. The dendrites receive synapses from other neurons, and the axon makes synapses on other neurons. So information enters a neuron through its dendrites, and leaves through the axon.

The starburst cell has no axon. In fact, that’s a defining characteristic of amacrine cells more generally. Instead, the starburst dendrites both receive synapses and make synapses on other neurons. (Perhaps we should refer to the dendrites as neurites, since the standard axon/dendrite distinction doesn’t hold, but that’s not the standard convention. In case you’re wondering, the J cell has an axon, like all ganglion cells, but the axon doesn’t show up in the overview because we didn’t map it.)


bouton EyeWire, neuron, 3D, synapseA part of a neurite that makes a synapse onto another neuron is called a presynaptic bouton. If you’ve helped to map the frist starburst cell (SAC), you may have already noticed the boutons. Those are locations where a dendrite swells for a short distance and then becomes thin again. Inside the swelling, you can seen that the dendrite is a bit darker than normal. This is probably due to the presence of vesicles of neurotransmitter, which are a defining characteristic of a presynaptic bouton.


If you look at the overview of SAC #1, you’ll see that the swellings are only in the distal regions of the dendrites, i.e., the swellings are far from the cell body and close to the tip.  So the distal dendrites of the starburst cell, instead of an axon, make synapses onto other neurons.


EyeWire SynapseSome of the boutons of the starburst neuron will contact J cells. Here is an image of one of them. EyeWire will begin to find first synaptic connections!


Over the next few weeks, we want to go much further. We would like to make a map of all starburst dendrites that are connected to J cells. At EyeWire HQ we are scouring the J dendrites to find points of contact with objects that look like starburst boutons. Each candidate starburst bouton will serve as a seed, and EyeWirers will trace where it goes. Some of these will turn out to be starburst cells, and some won’t. Let’s watch the overview together and see.  We don’t know what will emerge from our exploration.  That’s the excitement and suspense of doing science!


The mapping has several goals. Obviously, we’d like to know the numbers. How many starburst dendrites are connected to each J cell? How many synapses does each dendrite make on these J cells?  Less obviously, we’d like to know the directions of these dendrites. Are these directions uniformly distributed, or are they biased towards or against some direction?  To understand why this might be important for visual function, we will have to learn more about how starburst and J cells respond to visual stimuli.  Let’s cover that in the next post.

Leave a Reply

Your email address will not be published.