Research On Neural Circuitry Of Retina In Animal Models With Multi-electrode Array

The methods to detect the visual function of animals can be divided into two types:subjective method and objective method. The former one detects the visual functionmainly through behavior methods, such as training animals to identify target by foodinduction, and observing the head moving and the optokinetic nystagmus with aconvincing optomotor response phenomena and so on. However, it’s difficult to train theanimals and training period influents the result of subjective method. However, it’sdifficult to train the animals and training period influents the result of subjective method.At present, the objective method detects the visual function mainly through visualelectrophysiological examination technology, including electroretinogram (ERG), visualevoked potential (VEP), electrooculogram (EOG), etc. The technology can help toestimate the visual function objectively by recording the electrophysiological character o f the whole visual pathway from retina to visual cortex. But records of the technology arecomprehensive potentials. So the application is limited. For example, it can’t be used onresearch of single cell or cell group. Patch clamp can be used to record theelectrophysiological character of single cell, but it is difficult to research the relationshipbetween cells. For a neural network, the function of a single cell is based on the integrityof neuronal circuits. So the application of the technology of patch clamp is also limited inneuroscience research.In recent years, as the development of the micro electrode technology,multi-electrode array (MEA) recording system has been used in the field of neurosciencewidely. As a non-invasive method, MEA can record the electrophysiologicalcharacteristics of neurons synchronously. The technology can not only help to detect thereaction of single neuron, but also help to analysis the relationship between neurons. Thisexperiment is aimed at researching on neural circuitry of retina in animal models withmulti-electrode array, and providing theoretical foundation for further understanding thepathogenesis of some retinal diseases and exploring effective therapy.Materials and MethodsRetinas were isolated from experimental animals, Sprague-Dawley (SD) rats,congenital stationary night blindness (CSNB) rats, Kunming mice and rapid retinaldegeneration (rdf) mice, and placed into the recording chamber with the ganglion celllayer facing the biochip electrode array. Then the field potentials of retinas and spikes ofganglion cells, which were induced by electrical stimulation and different light stimulus,were recorded with a MEA system (MED-64, Alphamed Sciences, Osaka, Japan). Duringrecording, retinas were perfused withaerated Ringer’s solutionat rate of1ml/min. Analogextracellular neuronal signals from64channels were AC amplified (×1000), sampled at20kHz and stored in computer for subsequent off-line analysis. All recordings weresubsequently subjected to off-line spike sorting and analysis using OfflineSorter (PlexonInc, USA) and Neuroexplorer (Nex Technologies, MA, USA) software. Results1. The typical light-evoked responses were successfully recorded. And the amplitudeof field potential became higher as the luminance increasing (<256cd/cm2). According tothe peristimulus time histograms (PSTHs) and raster plots of individual units, the RGCswere categorized as ON, OFF and ON-OFF.2. The results also showed that the off-light responses were obtained whenprolonging the duration of light stimulus. In the wide type SD rats, a lower amplitudenegative wave was obtained at the offset of the light stimulus immediately, while a higheramplitude positive wave in CSNB rats.3. Besides, the long-latency (10~50ms) responses of RGCs evoked by electricalcurrent were declined in CSNB rats in comparison with SD rats (P <0.05).4. In the same luminance (256cd/cm2), the amplitude of light-evoked fieldpotentials of retina in Kunming mice are varied according to the wavelength of the light(blue> green> white> red, P <0.05).5. The typical light-evoked responses were not recorded successfully in rdf mice.And most of recorded RGCs were not responded to light. However, a few RGCs fired atthe time of light on and off. And this type of RGCs is more sensitive to blue light. In wildtype Kunming mice study, we also found this type of RGCs.Conclusion1. There are some shortages of MEA in evaluating retinal function. However, as anew technology, MEA integrates the traditional full field ERG and some function of patchclamp, so it can make a more comprehensive evaluation of retina function from severalaspects bases on the integrity of neuronal circuits.2. At present, most of the scientists consist that the d-wave of ERG is related to thefunction of OFF-center bipolar cells. In the research of CSNB rats, the off-light response,which was similar to the d-wave of ERG, was different from the response in wild type SDrats. So we suspected that the OFF pathway of retina was affected in CSNB rats.3. Electrical stimulation of RGCs induces two kinds of responses. Spikes with longer latencies (10~50ms) are attributed to indirect stimulation by synaptictransmission of activities from outer retinal layers. The synaptic transmission of outerretinal layers was disorder in CSNB rats. And the long-latency responses of RGCs evokedby electrical current were declined in CSNB rats in comparison with SD rats. So it mayprovide some evidence to the origin of the long-latency responses.4. Because of the intrinsically photosensitive retinal ganglion cells (ipRGCs) weresensitive to blue light, we can identify ipRGCs by analyzing the spectral sensitivity ofganglion cells from its electrophysiological characters.