University of Sussex
Browse
Darnet, Lea.pdf (26.47 MB)

Strategies to encode information with glutamate release in synapses of the Danio Rerio visual system

Download (26.47 MB)
thesis
posted on 2023-06-09, 17:53 authored by Lea Darnet
I have used the fluorescent reporter iGlusnFR to observe glutamate release from bipolar cell (BC) terminals onto retinal ganglion cells (RGCs) dendrites and from RGCs outputs in live zebrafish using multiphoton microscopy. Most neurons in the brain represent information using a digital code: temporal sequences of spikes of fixed amplitude that trigger the quantized release of neurotransmitter. The amplitude distribution of BCs events demonstrated clear quantization, showing that bipolar cells generate multivesicular events in vivo to encode visual information. I showed then that the vesicles constituting the events were released in a coordinated fashion and was not described by a Poisson process. It was then possible to understand for the first time how visual information was encoded with a vesicle code. Coding with amplitude was more prevalent in OFF cells than ON cells. Multivesicular events encoded higher contrasts with elevated temporal precision, achieving an accuracy comparable to spikes leaving the retina (about 3 ms). Ribbon synapses therefore discretize their outputs into sequences of numbers ranging from zero up to ~11 enhancing the dynamic range and the temporal accuracy of the vesicle code. Further, when observing iGluSnFR signals on the dendrites of individual RGCs, multiple individual inputs could be distinguished with varying sensitivity to tuning to spatial orientation. Thus, I used iGluSnFR to understand how visual information was transmitted onto RGCs by comparing inputs from BCs into single RGCs and outputs from the same RGCs in the optic tectum. I used this optical approach to study retinal computations such as dynamic predictive coding. Dynamic predictive coding is computed by RGCs of zebrafish. A combination of excitatory inputs from BCs and inhibitory inputs generate this phenomenon.

History

File Version

  • Published version

Pages

189.0

Department affiliated with

  • BSMS Neuroscience Theses

Qualification level

  • doctoral

Qualification name

  • phd

Language

  • eng

Institution

University of Sussex

Full text available

  • Yes

Legacy Posted Date

2019-05-21

Usage metrics

    University of Sussex (Theses)

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC