Combining two photon microscopy and CMOS MEA to uncover the mechanism of aberrant activity in blind retina

Abstract

Event Abstract Back to Event Combining two photon microscopy and CMOS MEA to uncover the mechanism of aberrant activity in blind retina Meng-Jung Lee1, 2, 3*, Prerna Srivastava3, 4, 5, Luke Rogerson3, 4, 5, Philipp Berens4, 5, 6, Thomas Euler4, 5, 6, Timm Schubert4, 5 and Günther Zeck1 1 Natural and Medical Sciences Institute, Neurophysics, Germany 2 Multi Channel Systems MCS GmbH, Germany 3 Centre de formation des diplômés en neuroscience, Universität Tübingen, Germany 4 Werner Reichardt Centrum für Integrative Neurowissenschaften (CIN), Universität Tübingen, Germany 5 Institut für Ophthalmologische Forschung, Universität Tübingen, Germany 6 Bernstein Zentrum für Computational Neuroscience Tübingen, Germany Motivation Increasing spontaneous activity and intrinsic oscillations have been reported in different cell layers in photoreceptor degenerated retina [1, 2]. Understanding the biophysical characteristics and the cellular origin of these oscillations may facilitate the development of strategies to treat diseases like Retinitis Pigmentosa and age-related macular degeneration. Here, we present the simultaneous recording of somatic calcium/synaptic glutamate signals in the inner and outer retina using two photon imaging and of action potentials of retinal ganglion cells (RGCs) in the ganglion cell layer using high-density CMOS based microelectrode arrays (MEAs). Material and Methods In this study, retinas from adult photoreceptor-degenerated rd10 mice were used. In the calcium imaging experiments, bipolar cells and amacrine cells in the retinal whole mount were labeled with Oregon Green BAPTA-1 using bulk electroporation [3]. For the glutamate imaging experiments, the AAV-encoded iGluSnFR biosensor was injected intravitreally into the mouse eye 3-4 weeks prior to the dissection of the retina [4].The labeled retina was placed RGCs down onto a poly-lysine coated CMOS-based MEA (CMOS 5000, Multi Channel Systems MCS) with a 1 mm2 sensitive area. The two photon recording was performed in a local area of 71.5 x 26.8 µm xy-scans for calcium imaging and 47.6 x 11.9 µm for glutamate imaging at different planes using an excitation laser tuned to 927 nm (Mai Tai DeepSee HP; Spectra Psychics). Data acquired from two photon imaging was analyzed with IGOR Pro, while spikes from RGCs were first sorted with CMOS-MEA-Tools and then further analyzed with Matlab. Results The main challenge of simultaneous two-photon imaging and extracellular recording using MEAs is the electrical artifact caused by the scanning laser beam in the sensitive electronic structures. This artifact occurred only in the MEA electrodes underneath the scanning laser. It could not be removed simply by applying filters to the signals as in previous work [5] . Instead, we applied a reset paradigm where the floating gate voltages of all sensing transistors were briefly assigned to one common voltage using internal switches. The reset parameters (duration and frequency) are defined by the user. Using this paradigm, continuous simultaneous two photon imaging and electrophysiological recordings with only 20 ms data loss during the resets were possible, yielding simultaneously calcium or glutamate signals in the outer and inner retina as well as spikes from the RGCs. Thus, combining two photon imaging and MEA recordings enabled us monitoring the activity in most retinal cell classes (remnant cones, bipolar cells, amacrine cells and RGCs) simultaneously. Conclusion Our results demonstrate that two photon imaging can be performed together with CMOS-MEA based recordings to study vertical signaling in the degeneratedive retina. This combination balances strengths and weaknesses of both recording techniques, offering a more complete and comprehensive way to study neuronal networks. Acknowledgements This work was supported by the Marie Curie Program switchBoard to M-J.L , P.S , T.S. and T.E. switchBoard receives funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie grant agreement No 674901. In addition, this work was funded by the Funding from the BMBF (FKZ031L0059A to G.Z. and FKZ 01GQ1601 to P.B.) is acknowledged by G.Z, the BW-Stiftung (NEU013) and the Deutsche Forschungsgemeinschaft (DFG, SFB 1233; BE5601/4-1).