The temporal structure of the inner retina at a single glance

Abstract

The retina decomposes visual stimuli into parallel channels that encode different features of the visual environment. Central to this computation is the synaptic processing in a dense layer of neuropil, the so-called inner plexiform layer (IPL). Here, different types of bipolar cells stratifying at distinct depths relay the excitatory feedforward drive from photoreceptors to amacrine and ganglion cells. Current experimental techniques for studying processing in the IPL do not allow imaging the entire IPL simultaneously in the intact tissue. Here, we extend a two-photon microscope with an electrically tunable lens allowing us to obtain optical vertical slices of the IPL, which provide a complete picture of the response diversity of bipolar cells at a "single glance". The nature of these axial recordings additionally allowed us to isolate and investigate batch effects, i.e. inter-experimental variations resulting in systematic differences in response speed. As a proof of principle, we developed a simple model that disentangles biological from experimental causes of variability and allowed us to recover the characteristic gradient of response speeds across the IPL with higher precision than before. Our new framework will make it possible to study the computations performed in the central synaptic layer of the retina more efficiently. The primary excitatory pathway of the mouse retina consists of photoreceptors, bipolar cells (BCs) and retinal ganglion cells (RGCs) (reviewed in refs. 1,2). At the core of this pathway is the inner plexiform layer (IPL), a dense synaptic plexus composed of the axon terminals of BCs, the neurites of amacrine cells, as well as the dendrites of RGCs. Specifically, the photoreceptor signal is relayed by the BCs to the RGCs via glutamatergic synapses (reviewed in ref. 3). This "vertical" transmission is shaped by mostly inhibitory interactions with amacrine cells, which integrate signals laterally along and/or vertically across the IPL (reviewed in ref. 4). Amacrine cells modulate, for instance, the sensitivity of BCs to certain spatio-temporal features 5-7. Within the IPL, the axon terminals of each of the 14 BC types 8-12 project to a distinct depth with axonal profiles of different BC types partially overlapping and jointly covering the whole depth of the IPL 10,11,13. Functionally, each BC type constitutes a particular feature channel, with certain temporal dynamics 7 , including On and Off BC types sensitive to light increments or decrements, respectively 14 , different kinetics 15,16 , and chromatic signals 17,18. Some of these features are systematically mapped across the IPL: For example, On BCs project to the inner and Off BCs to the outer portion of the IPL 14,19. Also kinetic response properties appear to be mapped, with the axonal profiles of more transient BCs localised in the IPL centre 7,15,20,21. To study BC function, early studies mostly used single-cell electrical recordings in vertical slices, where many lateral connections (e.g. large-scale amacrine cells) are severed, o...

Neural circuits in the mouse retina support color vision in the upper visual field

“…Like before, we defined ROIs based on local image correlation ( Fig. 3b; Methods) 35 . To register the IPL depth of each ROI, we used the two characteristic dendritic plexi of cholinergic starburst amacrine cells as landmarks (cf.…”

“…4f in ref. 35 ). In addition, we did not observe a bias for purely green center responses in the Off sublamina of dorsal scan fields, as would be expected for M-cone preferring type 1 BCs 30,55 .…”

System Identification with Biophysical Constraints: A Circuit Model of the Inner Retina

“…The first was a loss on the variance of the biphasic kernel speeds. Without this, we observed speed difference of up to 3x which is not biologically plausible (the biological range being (1 − 1.5) [34]). Secondly, we also minimized the variance of the mean and variance of the release output (after the ribbon block) for each BC type.…”

“…This allowed quantifying response consistency during the time of stimulus presentation. We recorded BC responses to these movies in n=1 Pvalb Cre mouse (Jackson laboratory, JAX 008069) using two-photon glutamate imaging in vertical optical slices (64x56 pixels @11.16 Hz) of the IPL, as described previously [34,5]. All animal procedures were approved by the governmental review board (Regierungspräsidium Tübingen, Baden-Württemberg, Konrad-Adenauer-Str.…”

Bayesian inference for biophysical neuron models enables stimulus optimization for retinal neuroprosthetics

“…Traces were recorded in one transgenic mouse (B6;129S6-Chat tm2(cre)Lowl J, JAX 006410, crossbred with Gt(ROSA)26Sor tm9(CAG-tdTomato)Hze , JAX 007905) that expressed the glutamate biosensor iGluSnFR ubiquitously across all retinal layers after intravitreal injection of the viral vector AAV2.7m8.hSyb.iGluSnFR (provided by D. Dalkara, Institut de la Vision, Paris). Cone glutamate release in the outer plexiform layer was recorded in x-z scans (64x56 pixels at 11.16 Hz; 63 ). Regions-of-interest (ROIs) were drawn manually and traces of single ROIs were then normalized and upsampled to 500 Hz as described previously 19,64 .…”