Receptive field organization of motion computation in the fly: a study of cell types and their variability

by Luis Giordano Ramos Traslosheros Lopez
Doctoral thesis
Date of Examination:2019-12-03
Date of issue:2020-12-02
Advisor:Prof. Dr. Marion Silies
Referee:Prof. Dr. Marion Silies
Referee:Prof. Dr. Tim Gollisch
Referee:Prof. Dr. Fred Wolf
Persistent Address: http://dx.doi.org/10.53846/goediss-8344

 

 

Files in this item

Name:RamosTrashlosheros-PhDthesis.pdf
Size:26.1Mb
Format:PDF
Description:PhD thesis

The following license files are associated with this item:

Abstract

English

It is believed that knowing all cell types making up a brain will lead to its understanding. Consequently, current efforts focus on classifying cell types according to their anatomical, genetic and physiological properties. Research in Drosophila melanogaster has greatly advanced cell type classifications, using transcriptomics, connectomics and genetic tools for neural circuit dissection. In particular, the fly eye, or optic lobe, has been extensively researched in an attempt to understand motion computation. Detection of motion is crucial for the survival of many species. Furthermore, motion computation is close to being mapped to its algorithmic neuronal substrate. In this thesis, we map the functional circuit organization of the fly visual system at the level of cell types. We focus on Tm9, the main synaptic input to T5 neurons, the first direction-selective neurons in the fly visual system, which detect moving dark edges. Tm9 like most neurons in Drosophila melanogaster has a stereotyped anatomy, that distributes regularly over the 800 columns of the fly eye and can be specifically targeted with genetic lines. Here, using in vivo two-photon calcium imaging we show Tm9 has variable receptive field properties that contrast with its homogeneous anatomy and genetics. In particular, Tm9 can display both narrow and wide receptive fields. We demonstrate this variability is not common to other fly neurons using simultaneous dual imaging, and find that wide-field responses are mediated by ON signals. Using anatomical and optogenetic circuit mapping, we identify Dm4, Dm12, and Dm20 as novel wide-field (variable) inputs to Tm9, and combine genetic silencing with in vivo imaging to show each of them mediates Tm9 response properties. In particular, Dm12 and Dm20 sharpen Tm9 ON receptive fields. The influence of Dm4, Dm12, Dm20, and Tm9 extends to downstream computations, genetic silencing reveals they are all required for proper directional tuning of T4 and T5 neurons. We further establish a computational role of functional variability in Tm9 by showing that Tm9 axons’ spatiotemporal properties correlate with the ones from postsynaptic T5 dendrites. Taken together our results unravel the existence of variability within classical cell types, its functional relevance in motion detection, and show newly characterized wide-field neurons in the fly brain are required for motion computation. We postulate flexible synaptic connectivity as a mechanism to regulate functional heterogeneity in morphologically- and genetically-defined cell types.
Keywords: motion detection; fly vision; receptive field; direction selectivity; neuronal circuits; wide-field neurons; cell types; calcium imaging; optic lobe; variability within cell types
 
Statistik