How does our ability to detect motion develop and to what extent does sensory experience contribute? In mice, the detection of motion begins in the retina. There, direction selective ganglion cells (DSGCs) fire more action potentials in response to visual stimuli moving in one direction, called the preferred direction, than visual stimuli moving in the opposite direction, called the null direction. In the adult retina, the preferred directions of DSGCs cluster along 4 directions. The relative orientation of these clusters varies across the surface of the retina, creating a direction selectivity map. The factors which underlie the development of the direction selectivity map is unknown. Two recent experiment indicate visual experience is involved. The first is that the preferred direction of DSGCs cluster along the axes defined by optic flow, i.e. the trajectories of apparent motion mapped onto the retinal surface. The second is from the Feller lab showing that the clustering of DSGCs preferred directions along the 4 axes requires visual experience.
In Aim1, I present strong preliminary evidence that demonstrates that the map is established at eye-opening and is not altered after visual deprivation, contradicting a role for activity. Although visual experience does not seem necessary for establishing the direction selectivity map, it may play a role in refining it. For example, raising a mouse in an artificial environment that enriches for specific orientations leads to an overrepresentation of cortical cells that prefer that orientation and the process is thought to be instructive rather than permissive. Therefore, in specific aim 2, I will determine whether altered visual experience, specifically enriched for optic flow, refines the direction selectivity map. Finally, in Aim 3, we explore the factors that influence the organization of the direction selectivity map at eye opening. First, we test whether retinal waves, which are spontaneous depolarizations of retinal ganglion cells that occur exclusively during the first two weeks of postnatal life, play a role. These waves are the primary driver of activity before eye opening in the direction selectivity circuit and even exhibit a propagation bias toward forward optic flow. We propose a pharmacogenetic experiments to determine whether disruption of retinal waves alters the organization of the map. Second, we begin to explore the role of transient molecular gradients that are expressed in the retina when the direction selectivity circuit is developing. Specifically, I provide evidence that the EphB receptor of the ephrinB axon guidance molecule exhibits a similar distribution gradient as the dorsal- ventral DSGC preferred directions. We will then look at maps in a mouse lacking EphB receptors in the retina.
The detection of motion begins in the retina, yet little is known about how the neural circuit that underlies this computation develops. The work described in this proposal will determine how visual experience, spontaneous activity, and transient molecular gradients help develop the retina's ability to detect motion with directional specificity. By understanding the strategies that the developing organism employs to build neural circuits, we aim to inform novel therapeutic tools for building functional sensory maps.
