Abstract:The mammalian AII retinal amacrine cell is a narrow-field, multistratified glycinergic neuron best known for its role in collecting scotopic signals from rod bipolar cells and distributing them to ON and OFF cone pathways in a crossover network via a combination of inhibitory synapses and heterocellular AII::ON cone bipolar cell gap junctions. Long considered a simple cell, a full connectomics analysis shows that AII cells possess the most complex interaction repertoire of any known vertebrate neuron, contacting at least 28 different cell classes, including every class of retinal bipolar cell. Beyond its basic role in distributing rod signals to cone pathways, the AII cell may also mediate narrow-field feedback and feedforward inhibition for the photopic OFF channel, photopic ON-OFF inhibitory crossover signaling, and serves as a nexus for a collection of inhibitory networks arising from cone pathways that likely negotiate fast switching between cone and rod vision. Further analysis of the complete synaptic counts for five AII cells shows that (1) synaptic sampling is normalized for anatomic target encounter rates; (2) qualitative targeting is specific and apparently errorless; and (3) that AII cells strongly differentiate partner cohorts by synaptic and/or coupling weights. The AII network is a dense hub connecting all primary retinal excitatory channels via precisely weighted drive and specific polarities. Homologs of AII amacrine cells have yet to be identified in non-mammalians, but we propose that such homologs should be narrow-field glycinergic amacrine cells driving photopic ON-OFF crossover via heterocellular coupling with ON cone bipolar cells and glycinergic synapses on OFF cone bipolar cells. The specific evolutionary event creating the mammalian AII scotopic-photopic hub would then simply be the emergence of large numbers of pure rod bipolar cells.
Abstract: An implantable retinal prosthesis has been developed to restore vision to patients who have been blinded by degenerative diseases that destroy photoreceptors. By electrically stimulating the surviving retinal cells, the damaged photoreceptors may be bypassed and limited vision can be restored. While this has been shown to restore partial vision, the understanding of how cells react to this systematic electrical stimulation is largely unknown. Better predictive models and a deeper understanding of neural responses to electrical stimulation is necessary for designing a successful prosthesis. In this work, a computational model of an epi-retinal implant was built and simulated, spanning multiple spatial scales, including a large-scale model of the retina and implant electronics, as well as underlying neuronal networks.
Three technologies have emerged as therapies to restore light sensing to profoundly blind patients suffering from late-stage retinal degenerations: (1) retinal prosthetics, (2) optogenetics, and (3) chemical photoswitches. Prosthetics are the most mature and the only approach in clinical practice. Prosthetic implants require complex surgical intervention and provide only limited visual resolution but can potentially restore navigational ability to many blind patients. Optogenetics uses viral delivery of type 1 opsin genes from prokaryotes or eukaryote algae to restore light responses in survivor neurons. Targeting and expression remain major problems, but are potentially soluble. Importantly, optogenetics could provide the ultimate in high-resolution vision due to the long persistence of gene expression achieved in animal models. Nevertheless, optogenetics remains challenging to implement in human eyes with large volumes, complex disease progression, and physical barriers to viral penetration. Now, a new generation of photochromic ligands or chemical photoswitches (azobenzene-quaternary ammonium derivatives) can be injected into a degenerated mouse eye and, in minutes to hours, activate light responses in neurons. These photoswitches offer the potential for rapidly and reversibly screening the vision restoration expected in an individual patient. Chemical photoswitch variants that persist in the cell membrane could make them a simple therapy of choice, with resolution and sensitivity equivalent to optogenetics approaches. A major complexity in treating retinal degenerations is retinal remodeling: pathologic network rewiring, molecular reprogramming, and cell death that compromise signaling in the surviving retina. Remodeling forces a choice between upstream and downstream targeting, each engaging different benefits and defects. Prosthetics and optogenetics can be implemented in either mode, but the use of chemical photoswitches is currently limited to downstream implementations. Even so, given the high density of human foveal ganglion cells, the ultimate chemical photoswitch treatment could deliver cost-effective, high-resolution vision for the blind.
Abstract: Classification and detection of biological structures in Electron Micrographs (EM) is a relatively new large scale image analysis problem. The primary challenges are in modeling diverse visual characteristics and development of scalable techniques. In this paper we propose novel methods for synapse detection and localization, an important problem in connectomics. We first propose an attribute based descriptor for characterizing synaptic junctions. These descriptors are task specific, low dimensional and can be scaled across large image sizes. Subsequently, techniques for fast localization of these junctions are proposed. Experimental results on images acquired from a mammalian retinal tissue compare favorably with state of the art descriptors used for object detection.
Connectomics is a strategy for mapping complex neural networks based on high-speed automated electron optical imaging, computational assembly of neural data volumes, web-based navigational tools to explore 1012-1015 byte (terabyte to petabyte) image volumes, and annotation and markup tools to convert images into rich networks with cellular metadata. These collections of network data and associated metadata, analyzed using tools from graph theory and classification theory, can be merged with classical systems theory, giving a more completely parameterized view of how biologic information processing systems are implemented in retina and brain. Networks have two separable features: topology and connection attributes. The first findings from connectomics strongly validate the idea that the topologies complete retinal networks are far more complex than the simple schematics that emerged from classical anatomy. In particular, connectomics has permitted an aggressive refactoring of the retinal inner plexiform layer, demonstrating that network function cannot be simply inferred from stratification; exposing the complex geometric rules for inserting different cells into a shared network; revealing unexpected bidirectional signaling pathways between mammalian rod and cone systems; documenting selective feedforward systems, novel candidate signaling architectures, new coupling motifs, and the highly complex architecture of the mammalian AII amacrine cell. This is but the beginning, as the underlying principles of connectomics are readily transferrable to non-neural cell complexes and provide new contexts for assessing intercellular communication.