Previous studies suggested that a glutamatergic-purinergic signaling pathway prevents hypoosmotic swelling of Müller cells in the rodent retina FK228 solubility dmso by vesicular release of glutamate (Wurm et al., 2008 and Wurm

et al., 2010; Figure 4A). Therefore, we tested whether this pathway was defect in BoNT/B-expressing Müller cells. Müller cells from Tam-injected bigenic mice, which expressed the BoNT/B transgene as indicated by the presence of EGFP (Figure 4B), had similar cross sectional areas as cells from Tam-injected monogenic mice (Figure 4C). However, these cells swelled in hypotonic solution, while Müller cells from monogenic animals maintained their cell volume (Figure 4D) indicating a toxin-induced defect in volume regulation. If the Quisinostat swelling of toxin-expressing Müller cells was due to the block of glutamate release, coapplication of glutamate should prevent the swelling of these cells. As shown in Figure 4D, this

was indeed the case thus confirming the involvement of exocytotic glutamate release in volume regulation. Vascular endothelial growth factor (VEGF), which activates the volume-regulating cascade upstream of glutamate release (Wurm et al., 2008), failed to abolish swelling of Müller cells from bigenic mice (Figure 4D), whereas it prevented hypotonic swelling of Müller cells from monogenic mice due to barium-induced block of inwardly rectifying potassium channels (Wurm et al., 2008) (Figure 4D). Toxin expression in Müller cells may have provoked reactive gliosis and thereby perturbed glial volume regulation (Pannicke et al., 2004, Pannicke et al.,

2005 and Sene et al., 2009). However, levels of glial fibrillary acidic protein (GFAP) were similar in retinae from Tam-injected mono- and bigenic mice (data not shown). Additional hallmarks of Müller cell reactivity are a downregulation of potassium inward currents and an increase in membrane capacitance (Pannicke et al., 2005). Our patch-clamp recordings revealed comparable amplitudes of inward currents and even a significant (p < 0.02; Student's t test) decrease in membrane capacitance in acutely isolated Müller cells from Tam-injected bigenic (2.4 ± 0.7 nA; 40 ± 12 pF; n = 33 cells from 8 mice) compared to monogenic mice (2.2 ± 0.6 nA; 49 ± 12 pF; Chlormezanone n = 29 cells, 7 mice). The VEGF-induced glutamate release from Müller cells also induces swelling of neurons in the ganglion cell layer independently from a hypo-osmotic challenge (Wurm et al., 2008). To test whether this effect was eliminated by glial toxin expression, we measured the size of neuronal somata that were close to endfeet of toxin- and EGFP-expressing Müller cells from Tam-injected bigenic mice (Figure 4E). Indeed, these neurons did not show VEGF-induced swelling, whereas the effect was present in cells from Tam-injected monogenic animals (Figure 4E, F). As for Müller cells, the mean cross sectional area of untreated neurons was similar in mono- and bigenic mice (Figure 4C).