ummaryPrefaceSummary (English)Summary (Dutch)List of Abbreviations Table of contentsIntroduction:Chapter1: A Projection specific logic to sampling visual inputs in mouse superior colliculus Chapter2: Characterization of GABAergic neurons in the mouse superior colliculusChapter3: Visual inputs to wide-field neuronsGeneral method ConclusionAppendicesCurriculum vitae (optional)List of publications (optional)ref: https://set.kuleuven.be/phd/researchers/template_kaft/guidelines_PhDthesis.htmSUMMARYInstinctive defensive behaviors, such as escape or freezing, are essential for an animal's survival. Sensory stimuli that innately represent threats are detected and conveyed to specific circuits that pass through brain regions. However, the rules by which parallel inputs are routed to downstream targets are poorly understood. The mouse superior colliculus mediates a set of visually guided innate behaviors by receiving input from >30 retinal ganglion cell types and projecting to behaviorally important circuits. These circuits are mediated by several groups of collicular neurons with diverse morphology, visual response properties, and long-range targets including the pulvinar, lateral geniculate nucleus, and parabigeminal nucleus. Each collicular neuron has been estimated to receive input from on average six retinal ganglion cells. However, the different ganglion cell types that provide input to specific superior collicular output pathways have not been characterized. By comparing the morphological, molecular and visual response properties of retinal ganglion cells innervating colliculo-parabigeminal and colliculo-pulvinar pathways through the superior colliculus, we found a projection-specific logic where each collicular output pathway sampled a distinct set of retinal inputs. These findings raise the possibility that projection-specific sampling of retinal inputs forms a basis for the selective triggering of behaviors by the superior colliculus.To test this idea, we focus on a population of collicular neurons - horizontal cells, which provide inhibitory inputs to downstream targets, including the lateral geniculate nucleus and parabigeminal nucleus. We ask whether each output pathway shares a common set of retinal inputs and whether the different pathways mediate distinct behaviors. Using a retrograde tracer and characterization of the neuronal response properties, we found that horizontal cells are made up of two anatomically and functionally distinct populations defined by their projections to LGN and Pbg. Manipulating these two distinct populations that target either LGN and PBg results in different behavior outputs. By comparing the ganglion cell types providing inputs to these circuits. We found xxx. Token together, our findings strongly support the notion that, in the superior colliculus, neuronal circuits are based on a dedicated set of connections between specific retinal inputs and different collicular output pathways. INTRODUCTIONAnimals, including humans, rely mostly on the vision to explore the world, make decisions, and perform various behavioral tasks. Visual processing starts from the retina and is carried out by distinct cell classes and their neurites that are located at segregating retinal layers, which results in a number of unique visual features. The output neurons, retinal ganglion cells (RGCs) consist of more than 30 functionally defined types (Baden et al 2016), each responds best to a particular feature of the visual world. They directly send parallel visual information to multiple brain targets and further to trigger a series of different behaviors through other downstream circuits. Among the diverse central projections of the retinal ganglion cells, the superior colliculus (SC), as well as its non-mammalian homolog, the optic tectum, is an evolutionarily conserved brain structure across different animal species. The SC is a laminar structure, located at the dorsal midbrain. In mice, the SC receives more than 85-90% of direct retinal input from retinal ganglion cells (Ellis et al., 2016), which makes it the most prominent visual center in this species. More than five cell types of SC neurons have been identified based on their morphology and physical response properties. Activation of the specific SC cell types mediated neuronal circuits has been shown to trigger different behaviors, such as freezing and escape. 1. Mouse visual system1. 1 Visual processing starts from the retinaThe mouse retina is a laminar structure consisting of five cell classes of neurons, which could be further divided into distinct cell types. They are assembled and organized into distinct circuits that enable visual processing. Photoreceptors including rods (daylight detection) and L- and S-cones (nighttime vision), reside in the outer nuclear layer (ONL) of the retina. They convert the light information into electrical signals. Horizontal, bipolar, and amacrine cells are the interneurons that process the output of photoreceptors and occupy the inner nuclear layer (INL) of the retina. There is one single type of horizontal cells and at least 13 types of bipolar cells in the mouse (Euler et al. 2014; Shekhar et al 2016). Horizontal cells provide lateral feedback inhibition to regulate glutamate release from the photoreceptors. Bipolar cells receive glutamatergic inputs from photoreceptors and GABAergic inputs from horizontal cells from the outer plexiform layer (OPL). In the inner plexiform layer (IPL), bipolar cells relay their signals to amacrine cells and converge in specific ratios onto the dendrites of retinal ganglion cells (RGCs). Amacrine cells can be classified broadly as narrow or wide-field on the basis of the diameters of their dendritic trees. They make inhibitory synapses onto bipolar cell axon terminals and onto the dendrites of other amacrine and ganglion cells (Demb & Singer 2015). The cell bodies of RGCs reside in the ganglion cell layer (GCL), the innermost nuclear layer. The RGCs receive the filtered and shaped signals from bipolar and amacrine cells and convey visual information to the rest of the brain through the optic nerve (Fig. 1).