We next investigated whether toxin expression in

Müller cells affects retinal function at different levels of integration. First, we used multielectrode arrays (MEAs) to record spike activity in the retinal ganglion cell layer in response to visual stimuli ex vivo (Figure 5). Previous studies showed that glial glutamate release inhibits light-evoked ganglion cell activity (Newman and Zahs, 1998). We stimulated retinae with different monochromatic visual stimuli including fullfield flashes, drifting bars, and drifting gratings to explore the spatial and temporal response properties of ganglion cells. Ganglion cells from adult Tam-injected mono- and bigenic mice did not differ in their responses to simple steps in light intensity (Figure 5A) find more RO4929097 or to sinusoidal drifting

gratings of various temporal and spatial frequency and contrast (Figure 5B). Ganglion cells show accelerated response kinetics in a high-contrast environment (contrast adaptation; Baccus and Meister, 2002). We also observed this phenomenon (Figure 5C), and there was no difference between bi- and monogenic mice. Finally, we measured the velocity tuning of ganglion cell spikes in response to bars drifting across the receptive field at various velocities (Figure 5D). Again, there was no difference between the two mouse lines. Second, we recorded electroretinograms (ERGs) to measure light-evoked electrical responses of retinal layers in vivo using single flash and flicker stimuli (Tanimoto et al., 2009; Figure 6). However, our experiments revealed no detectable differences in the retinal responses of Tam-injected

bigenic mice compared to monogenic littermates (Figure 6). Third, we performed behavioral tests to assess visual function in mice (Figure 7; Arqué et al., 2008). For these experiments, we used Tam-injected below bi- (n = 12–13) or monogenic (n = 12–17) males. In the novel object recognition (NOR) test, Tam-injected bi- and monogenic mice spent more time exploring the novel object compared to the old one, but the mean fraction of time was similar in both groups (Figure 7A). In the water maze test with a visible platform, bi- and monogenic mice found the visible platform with similar latencies (Figure 7B) and in the open field test, mice from both groups spent similar times in the central zone (Figure 7C). Finally, we tested whether toxin expression in glial cells affects the photic entrainment of the circadian rhythm, which is mediated by photoreceptors and light-sensitive cells in the ganglion cell layer (Golombek and Rosenstein, 2010). To this end, we recorded running-wheel activity of bi- and monogenic mice during a 12/12 hr light/dark cycle before and after Tam-induced toxin activity.