The nervous system contains a diverse array of neuronal and glial types which must coordinate their growth and signaling as they integrate into circuits. Among sensory systems, these neural circuits are tuned to extract salient environmental features and enable an organism to perceive and respond to stimuli. Reliable feature extraction is achieved via precise synaptic connections among inhibitory and excitatory neurons which enable computations to be performed on inputs. Glial cells play essential roles in the development and function of these circuits, for instance by releasing synaptic modulators or by limiting the spread of released neurotransmitters. The retina is an ideal model for studying the coordinated development of neural circuits and the roles glial cells play in this process.

This dissertation begins with a description of the current understanding of Müller glia development in Chapter I, starting with their specification from progenitors and ending with their integration into retinal circuits. Chapter II describes experiments aimed at defining mechanisms by which Müller glia respond to neuronal activity during development, and whether these responses modulate outgrowth of glial processes into retinal synaptic layers. I find that these processes undergo compartmentalized calcium transients in response to retinal activity while simultaneously undergoing rapid motility, but that this motility is calcium- and neuronal activity-independent. Although I do not identify a function for neuron-glia signaling in Müller glial motility, there is evidence that glia sculpt developing retinal circuits in a variety of other ways, perhaps through the action of neurosteroids. In Chapter III, I investigate the function of Trpm3, a neurosteroid receptor expressed in Müller glia and subsets of inner retinal neurons in the postnatal retina. I find that steroid activation of Trpm3 induces prolonged calcium influx and activation of immediate-early genes in retinal ganglion cells, along with enhanced activity presynaptic to retinal ganglion cells.

Up to this point, I have investigated generally how circuits in the retina function during development, and how they interact with signals associated with their glial cell constituents. In Chapter IV, I describe how a particular circuit tuned for motion direction develops, with a focus on secreted proteins which function as synaptic organizers. The cells that make up this circuit, the starburst amacrine cell and the direction-selective ganglion cell, form stereotyped connections which confer sensitivity to one of four cardinal motion directions for each direction-selective ganglion cell. I report a panel of candidate genes which may mediate this wiring precision, and I follow up on one of these molecules, cerebellin-4. I find that cerebellin-4 is dispensable in retinal ganglion cells for their precise subcellular wiring onto starburst cells, but that its absence is associated with weakened synaptic responses to light stimuli. Together, the results described in this dissertation highlight the many essential cell-cell interactions which occur within a small window of development to coordinate the formation of functional neural circuits.