BRAIN_WIRE · Functional and molecular characterization of excitatory layer IV neurons in mouse visual cortex

Functional and molecular characterization of excitatory layer IV neurons in mouse visual cortex

A universal hallmark of a sensory cortex is the selectivity of individual neurons for particular features of sensory scenes. Together, the activity of these neurons gives rise to the perception of external stimuli. How individual neurons acquire their selectivities during development remains an important unsolved question of systems neuroscience. For each neuron, response properties are largely defined by the pattern of inputs it receives. What role does the molecular makeup of individual neurons play in selecting their inputs and determining their functional properties? Are there subclasses of excitatory neurons preprogrammed to process specific classes of inputs? To begin to answer these questions this project set out to characterize the contribution of neuronal molecular identity to feature selectivity in the visual system. To accomplish this goal, we have developed a method for simultaneous functional and transcriptional profiling of single cortical neurons. We first characterize the visual selectivity of neurons in vivo using two-photon calcium imaging, including tuning to orientation, direction, spatial, and temporal frequency, as well as modulation by running and pupil diameter. We then identify the imaged neurons in acute brain slices, harvest the cellular contents by microaspiration, and measure gene expression by single cell RNA sequencing. We first confirmed that this transcriptional profiles of prepared using this approach could distinguish well-established classes of neurons in the cortex. We collected RNA samples from parvalbumin- and somatostatin-positive interneurons, identified in transgenic mice expressing tdTomato in these populations, and layer 2/3 and layer 5 pyramidal neurons in mouse V1. Expression of marker genes readily distinguished the transcriptional profiles of these cell-types. We then asked whether we can confirm known correlations between gene expression and neuronal responses. First, we used single-cell transcriptional profiles to identify pyramidal neurons and parvalbumin-positive interneurons and compared the orientation tuning of their visual responses. Consistent with what we know about the properties of parvalbumin neurons, they were broadly tuned to orientation. Second, we examined the in vivo activity patterns of excitatory neurons, expressing high levels of the immediate-early gene Fos. Cells expressing high levels of Fos displayed higher rates of calcium transients during visual stimulation in vivo. We have focused primarily on excitatory neurons in layer 2/3 of primary visual cortex. Although their gene expression patterns are heterogeneous, no clear subclasses of can be detected based on their transcriptional profiles alone. This does not eliminate the possibility that expression levels of small numbers of genes contribute to the diversity of orientation, direction, spatial and temporal frequency tuning of this population. However, hundreds of cells are needed to achieve the statistical power required to detect such a relationship and we are currently building up such a sample. This project, which we hope to complete within the next few months, will provide an important resource cataloguing the functional and molecular diversity of mouse visual cortex neurons.

Data: CORDIS, © European Union