Abstract:
The oculomotor integrator transforms an eye velocity input into an eye position signal that
is essential for retinal image stabilization. This signal is stored for a prolonged amount of
time in the brain and acts as a working memory for eye position. The oculomotor integrator
is embedded into the vertebrate hindbrain circuit that drives horizontal eye movements.
While the basic components of this circuit are known, we still have little knowledge about
how those neurons achieve this integration and the exact coding properties of those
neurons and of the connected oculomotor nuclei. Similarly, different theories exist that try
to explain how the binocular coordination of the two individual eyes is achieved to drive
precise binocular eye movements. No conclusive answer has been found to this question so
far.
The larval zebrafish is becoming an increasingly popular choice for neuroscientists as a
model organism due to its transparency and great accessibility for microscopy studies
during early development. Therefore, I used calcium imaging in the developing brain of
larval zebrafish to investigate those questions.
In the first set of experiments, I exploit a specific experimental paradigm to provoke
monocular eye movements in one-week old larval zebrafish to investigate the eye specific
tuning of oculomotor neurons and to coarsely map their eye position/velocity sensitivity. I
imaged the hindbrain area encompassing the nucleus abducens, oculomotor integrator,
inferior olive, and the velocity storage mechanism. The results of this experiment show
that the neurons of those nuclei can be grouped into four response groups which differ in
their activity during monocular eye movements. One group shows preferential activity
during binocular eye movements. This points towards a certain degree of task separation at
this developmental stage. Additionally, I show how the oculomotor integrator appears to
extend into areas that were previously not identified as important for retinal image
stabilization.
In the second part, I further investigate the precise tuning properties of those neurons by
running a closed-loop experiment that was aimed at decoupling the eye position signal
from eye velocity. I report how oculomotor neurons encode eye position and eye velocity
to a varying degree and their different activation thresholds. I show how the neurons in the
caudal hindbrain appear to integrate eye velocity into position along a gradient, but they
are arranged in two separate eye position and eye velocity clusters.
In the last set of experiments, I examine the maturation of the previously investigated
nuclei. I replicate the experiments on two-week old larvae, as the brains of zebrafish are
still developing at that age. I show how several aspects of the oculomotor system are
already established in young larvae, but some others are still undergoing refinement.
Monocular neurons increase their eye specific sensitivity even further and the eye velocity
system is becoming almost exclusively monocular with age. Neurons that show
preferential binocular activity are more distributed in the brain and become less frequent
with age, while neurons that are active regardless which eye is moving become more
abundant.
This thesis characterizes the binocular coordination and coding sensitivities of the
oculomotor hindbrain neurons at two different developmental stages. It expands our
knowledge on how the nuclei controlling horizontal eye movements are tuned and provides
the basis for further investigations on how persistent activity can be generated in the brain.
Print-on-Demand