Visualization of dopaminergic signaling in the retina with optogenetic sensors

The retina is able to adapt to light intensities that range over several orders of magnitude. Dopamine (DA) is a neuromodulator that is released by one amacrine cell type in the retina. In the early eighties, DA was already discussed to play a central role in light adaptation processes. DA exerts it effects via G-protein coupled receptors (GPCRs), which were grouped into two families: the D1-family (D1R), whose activation leads to an increase in adenylate cyclase (ACy) activity, and the D2-family (D2R), whose activation reduces the activity of the ACy. Thus, these DA-receptor (DR)-mediated pathways are coupled to the regulation of protein kinase A (PKA) activity. Besides this classical cyclic adenosine monophosphate (cAMP)/PKA cascade, multiple alternative pathways have been described which amongst others may lead to changes in the intracellular concentration of the central second messenger calcium ([Ca2+]i). Up to now, in the retina multiple effects of DA are known but the underlying mechanisms are still unresolved. To gain a better comprehension of these mechanisms, the present study was designed to investigate the regulation of DA release as well as the DA-mediated pathways in target cells. The mouse retina served as model system. To monitor the release of DA in the retina in a time-resolved manner, the genetically-encoded sensor synapto-pHluorin should specifically be expressed in dopaminergic amacrine cells. For test purposes, a construct was cloned in which the expression of the marker protein GFP (green fluorescent protein) was controlled by the tyrosine hydroxylase (TH) promoter. Expression of this construct in HEK293 cells and retinal neurons in culture revealed that the promoter did not exhibit the expected specificity: GFP was expressed in TH-negative cells. The impact of DA on the intracellular cAMP concentration ([cAMP]i) in target cells was investigated using the genetically-encoded Förster-resonance-energy-transfer (FRET)-based sensors EPAC1-camps and AKAR4. Initially, the two sensors were expressed in HEK293 cells and in cultured retinal neurons via Lipofectamine-transfection and were characterized. Both sensors made it possible to visualize production and degradation of cAMP after stimulation of GPCRs. Using a pharmacological approach, it was shown that the activation of D1R triggered an increase in PKA activity in a population of retinal neurons. In Ca2+-imaging experiments with Fluo-4-loaded cultured retinal neurons, it was demonstrated that DA alters [Ca2+]i in about 20% of cells. Different response types were found. Pharmacological investigations led to the assumption that the classical D1R/PKA-pathway as well as alternative pathways participate in DA-induced increases in [Ca2+]i. The DR-type mediating the DA-induced decrease in [Ca2+]i could not be identified, yet. In order to employ genetically-encoded sensors in the intact retina, the method of viral gene transfer was established. Adeno-associated viruses (AAVs) served as gene-shuttles. It was shown that AAV2 transduced neurons in vitro as well as in vivo that are regulated by DA. However, dopaminergic amacrine cells were never infected. In addition, viral expression of EPAC1-camps was not successful in retinal neurons although smaller sensors such as GCaMP3.0 could be functionally expressed with this method. Under the use of a transgenic mouse line that expresses the genetically-encoded Ca2+-sensor TN-L15 in ganglion cells (GC), it was shown that DA influences [Ca2+]i in ~50% of GCs. Different response types were found. Using pharmacological and fluorescence-microscopical methods it was proven that GCs of distinct response types differed morphologically (soma size) and physiologically (ON-/OFF-type). Selective blockade of the excitatory or inhibitory input to GCs indicated that a part of the DA-effect was based on a direct action of DA at the GC itself, while another part was based on the presynaptic effect of DA in the retinal network