Mystic Update: Zebrafish Progress Report

EM reconstructions of brainstem neurons. A. 3D rendering of reconstructed neurons. Large green cell body in the foreground is the Mauthner neuron (ro – rostral; c – caudal; d – dorsal; v – ventral; l – lateral; m – medial). Inset (top left) shows location of the unilateral EM volume (black box) relative to the olfactory bulb (OB), tectum (TE), hindbrain (HB), and spinal cord (SC). B-D. Automatic synapse detection and partner assignment. B) Raw EM image. Scale bar is 750nm. C) Postsynaptic densities (PSDs) identified by a convolutional net. D) PSDs (red) overlaid onto the original raw image, together with an exemplar presynaptic (blue) and postsynaptic (yellow) partnership identified by a second convolutional net. E. Sagittal view of identified abducens motor (ABDM, green) and abducens internuclear (ABDI, magenta) neurons overlaid over representative EM planes (R – rhombomere; * – Mauthner cell soma). F. Coronal planes showing the locations of the ABDM (left) and ABDI (right) neurons at the planes indicated by dotted black lines in e sagittal view. Black boxes highlight nerve bundles from these populations. G. Representative ABDM and ABDIneurons with arrows indicating the axons. H. Reconstructions of large and small reticulospinal (RS) neurons and dorsal octaval (DO) neurons.

The following is a zfish update from Seung Lab postdoc Ashwin Vishwanathan:

We began with a simple question, how does the brain keep the eyes still? We need to keep our eyes still or else we would have a very ‘shaky’ picture of the world. A simplistic way to think of how the eyes move is to think of a ball (the eye) which is fixed in place that can be turned from side to side by two eye-muscles. To move the eyes the eye-muscle needs to accurately apply just the right amount of force or else the eyes would over or undershoot. Once at its new location, the muscle needs to apply a constant force to keep the eye still at that location. We know that this is achieved by neurons in the brainstem that maintain a ‘memory’ of the eye location. This memory is encoded in the form of bursts of activity, more close to the nose the activity is low, farther away the activity is high that is subsequently passed on to the eye-muscles. Thus a recording of how ‘bursty’ the neurons are tells us where the eyes are. Although much of this was known for the last 50 years, the mechanism by which a network of neurons keep bursting at a constant rate was not known.

Theoretical work using simplistic models for neurons suggests that one way this could be possible is when neurons feed their activity back to other neurons in the same network, thus propagating the activity thereby keeping the eye-muscle contracted. Moving the eye to a new location would change the activity pattern from the existing pattern to get to the intended eye location. To do this, theory suggests that neurons in this network need to connect to each other in a specific pattern such that the activity can propagate in an orderly manner. However we can imagine that neurons in a network can be wired to each other in many different ways. Our goal for this project was to discover how the ‘real’ network is wired up.

To study this question we reconstructed the wiring diagram of the larval Zebrafish brainstem with the help of citizen scientists on Eyewire. We analyzed these reconstructions by dividing them into communities of neurons. We discover that one community was responsible for eye movements and the other for body movements, with a little cross talk between these communities. Furthermore, neurons responsible for eye movements  preferentially connected to other neurons in the same community. This was one of the main predictions from the theoretical work in the past. Finally, when we revisited the theoretical models but used the actual connectivity pattern from the reconstructions, we were able to precisely predict eye movement. Thus, we show that just connectivity alone can explain, in large part, activity patterns observed among neurons in a network. 

This work was a team effort and would not have been possible without the beautiful reconstructions from the Eyewire community. 

 

Read the paper here. 

For more info, check out this Twitter explainer.

 

A note from HQ: Mystic/Zfish zone of Eyewire would not have been possible without the brilliant Scouts Log implementations and updates from player and SL creator @eldendaf. We’re immensely grateful for his teamwork in this endeavor. We’re also grateful to each and every Mystic who helped make this a reality. Thank you – for science!