What the fish’s eye tells the fish’s brain: the functional complement of zebrafish retinal ganglion cells

In the vertebrate retina, incoming visual information is split into parallel information channels and sent to the brain by distinct types of retinal ganglion cells (RGCs) (Masland, 2012). In mouse, a minimum of 32 RGC types, and possibly as many as 50, have been charted functionally (Baden et al., 2016), anatomically (www.museum.eyewire.org; – see also RGCtypes.org) (see also Sanes and Masland, 2015) and transcriptomically (Bae et al., 2018). However, a similarly detailed understanding of the complement of zebrafish RGCs and their underlying feature-extracting microcircuits is lacking. Here, we performed in vivo two-photon population imaging of light-evoked calcium activity in the tetrachromatic zebrafish during hyperspectral visual stimulation, to functionally chart what the fish’s eye tells the fish’s brain. For this, we generated a novel transgenic line in which the calcium reporter GCaMP6f is tagged to the membrane of zebrafish RGCs under the Islet2b promoter, allowing recording from both RGC dendrites and somata. We find that RGC functional properties varied strongly with position in the eye, including a regional specialisation of UV-responsive Onsustained RGCs in the acute zone, likely to support visual prey capture. Interestingly half of RGCs display diverse forms of colour opponency, among which many are driven by a pervasive and slow blue-Off system, challenging classical models of colour vision. In addition, spectral contrast was intermixed with temporal information. Taken together, our data suggest a highly regionalized time-colour code that asymmetrically encodes distinct regions in visual space, mirroring our earlier finding in bipolar cells that different parts of the eye harbour strongly diverse functional circuits (Zimmermann et al. 2018). Working toward more ecologically and behaviourally relevant stimuli, we next used a novel prototype of a tetrachromatic spatial stimulator (Franke et al. 2019) to map receptive fields, and potentially make better sense of more complex contextual effects such as sensitivity to a certain orientation or direction of motion. Crucially, analysing receptive fields in a population of cells can provide insight into the visual features extracted at a particular stage along the visual pathway.