What the mouse eye tells the mouse brain: Fingerprinting the retinal ganglion cell types of the mouse retina

Abstract

The retinal ganglion cells (RGCs) relay the output of each parallel feature detecting channel established through complex interactions in the retinas two synaptic layers to higher visual centres. Understanding how the visual scenery is encoded by the outputs of the 20 RGC types will thus yield a complete picture of the representation of the visual scene available to the brain. To reliably record from each RGC type in the mouse retina, we bulk-electroporated the tissue with a synthetic calcium indicator (OGB-1) and used two-photon calcium imaging to record light stimulus-evoked activity at the level of the ganglion cell layer (GCL)(Briggman Euler, J Neurophysiol 2011). So far, our database contains recordings of> 10,000 cells from the GCL. In addition, we obtained recordings from transgenic PV and PCP2 mice, in which 13 morphologically distinct RGC types are fluorescently labelled and can be identified based on their anatomy (Farrow et al., Neuron 2013; Ivanova et al., J Comp Neuol 2013). Moreover, we performed electrical single-cell recordings from RGCs to relate their spiking responses to the somatic Ca2+ signals and to compare their morphologies with published RGC catalogues (eg, Völgyi et al., JCN 2009). We implemented a probabilistic clustering framework for separating RGCs into functional types based on features extracted from their responses to the different visual stimuli using PCA. We employed an automated mixture of Gaussians Clustering algorithm to cluster the cells based on their physiological properties. Subsequently, clusters were grouped according to genetic labels and morphological criteria (eg soma size). For our data …