Retinal ganglion cells: when and how do they contribute to the design and function of the developing visual system?

Lead Research Organisation: Newcastle University
Department Name: Institute of Neuroscience

Abstract

Understanding how neural function is generated is undoubtedly one of the most challenging current questions in neuroscience. One powerful approach to decipher neural function is to study how neural connections are formed in early life. Brain connectivity is initially coarse, determined by the genetic code and molecular signals. Later on, connections become refined under the influence of electrical activity which is ubiquitous in the developing central nervous system (CNS). It starts during gestation, long before sensory experience is even possible. It is spontaneously driven and takes distinct forms in different parts of the developing CNS. It is widely recognised that the precise spatial and temporal patterns encoded in early spontaneous activity are extremely important for guiding the refinement of neural connections.

We are interested in understanding how early neural activity guides the maturation of the visual system. Waves of spontaneous activity sweep across the layer of retinal ganglion cells (RGCs), the output channels of the retina during a limited perinatal period. Many studies have demonstrated that retinal waves guide the development of retinal function and the organisation of retinal central projections, principally the dorsal lateral geniculate nucleus (dLGN) of the thalamus and the superior colliculus (SC). In the immature brain, RGC projections from both eyes are intermingled and diffuse. With development, they segregate into eye-specific areas and become organised into precise topographic maps, with neighbouring RGCs projecting to adjacent cells in the dLGN and SC, resulting in the preservation of a coherent map of our visual world. Retinal waves play a crucial role during the process of eye-specific segregation and map refinement. However, previous studies have not considered that the dynamic properties of the waves profoundly change with development. We have recently shown in mouse (our experimental model) that waves are initially infrequent (hence unlikely to occur simultaneously in both eyes), very large and they recruit only a fraction of RGCs on their trajectory (many RGCs remain silent). Later on, they become more frequent, spatially smaller, slower and much denser, recruiting most neighbouring RGCs. Finally, shortly before the onset of visual experience, waves become almost stationary, turning into small and frequent hotspots that tile the retina. They completely disappear immediately after eye opening.

We hypothesise that early waves are important for eye-specific segregation (generating competition between the eyes), later waves are instrumental for the refinement of topographic maps (when waves become denser). In the latest period, waves are important for the maturation of retinal networks underlying the generation of light responses in RGCs, at a time when projections are mostly mature.

We will test this hypothesis using pharmacogenetics, a powerful approach allowing us to selectively and reversibly silence RGCs. For this purpose, we will generate a mouse line where most RGCs express Designer Receptors Exclusively Activated by Designer Drugs (DREADDs). When exposed to the designer drug clozapine n-oxide, DREADD-expressing RGCs are silenced, hence they do not participate in waves.
Using this approach, we will chronically silence RGCs during distinct perinatal periods, when waves convey different spatial and temporal information. Using electrophysiology, neuroanatomy and behaviour, we will investigate how silencing waves at different postnatal periods affects visual function in adults, focusing on:
1. Retinal connectivity and how RGCs respond to light.
2. Map refinement in the SC
3. Eye-specific segregation in the dLGN and SC
4. Behavioural measures of image details discrimination (visual acuity) and sensitivity to contrast

This project will elucidate how retinal waves guide different phases of the development of the visual system and how plastic the process is.

Technical Summary

In the visual system, spontaneous waves of activity sweep across the retinal ganglion cell (RGC) layer in perinatal life. These waves are known to guide eye-specific segregation and topographic map refinement in the dorsal lateral geniculate nucleus (dLGN) and the superior colliculus (SC). We recently reported that mouse retinal waves undergo profound spatiotemporal changes with development, becoming more frequent, narrower, slower and denser, recruiting more neighbouring RGCs on their trajectory. Finally, just before eye opening they become small activity hotspots, reminiscent of RGC receptive fields (RFs).

We hypothesise that early waves are important for eye-specific segregation; later on, when they become denser, they help refine topographic maps; finally, late waves guide the development of retinal RFs when projections are mostly mature.

To test this hypothesis, we will selectively and reversibly silence ~80% RGCs expressing the transcription factor Brn3b during distinct developmental periods using Designer Receptors Exclusively Activated by Designer Drugs (DREADDs). DREADD-Brn3b RGCs will be chronically silenced through systemic injections of the designer drug clozapine n-oxide. We will investigate how silencing waves that convey different spatiotemporal information influences:
1. Retinal connectivity and RGC RFs. This will be done with immunocytochemistry and in vitro recordings with a large-scale high density system (4096 channels) allowing us to record from thousands of RGCs across the entire retina.
2. Eye-specific segregation and map acuity in SC neurones. This will be achieved through in vivo recordings from the SC using Utah arrays with 25 probes.
3. Eye-specific segregation and RGC axonal arbours terminal zone organization visualized following intraocular injections of cholera-toxin subunit B-conjugated chromophores and DiI.
4. Behavioural measures of visual acuity and contrast sensitivity using the optokinetic tracking response.

Planned Impact

We identify the following categories that will benefit from this project:

ACADEMIA: Our work will generate high impact new knowledge on how circuits are formed in the CNS and how the visual system operates. It will benefit academics interested in the developing visual system, and in activity dependent refinement of brain connectivity in general. Theoreticians will have access to our data to develop and validate their models. Through the use of our very large and dense multielectrode array system for the retina and Utah array for in vivo recordings, we will generate extremely rich datasets that are valuable for engineering and bioinformatics.

PUBLIC SECTOR: the new knowledge we will generate to understand developmental plasticity will directly benefit the quality of academic teaching, starting with Newcastle, and also other institutions upon dissemination of our findings. Together with her collaborator B. Cessac, the PI will endeavour to develop new simulation tools deriving from the project to help teach neural development and plasticity and visual function. She is actively involved in the development of neuroscience teaching tools in African universities through her activities in the International Brain Research Organisation. These new simulation tools will be particularly beneficial for neuroscience teaching in developing countries, where students are less likely to be directly exposed to novel experimental findings.

BUSINESS/INDUSTRY: Considering the intense endeavours to develop therapeutic strategies to restore function following brain trauma or disease, our results will provide important information for neonatal medicine, with a particular emphasis on strategies to ensure healthy development in preterm babies. We will communicate our findings to clinicians and surgeons by presenting our work at the ARVO meeting, and by organising "Meet the Scientist" gatherings with clinicians and patients in the Royal Victoria Infirmary.
Our work will make use of cutting-edge recording technologies, providing a state-of-the-art environment to showcase how recordings from the retina using a large-scale, high-density system improves our understanding of how the retina functions. This will help the company that manufactures the system to attract new customers and develop new features for this sophisticated system.
"Big data" is becoming very important in different applications and the refinement of appropriate algorithms for the analysis of our rich datasets will benefit scientists working with similar approaches.
By training a highly skilled researcher and students who will join the project at various stages of their studies (undergraduate project students, master or PhD students) we will educate and form excellent researchers, ready to enter the workforce, not only in academia, but also in the Industry sector, and therefore contributing to the UK economy.

GENERAL PUBLIC: Vision is probably our most cherished sense, and is therefore well-suited to be understood by the general public. We will disseminate our work through public lectures and other activities such as café scientifique or science fairs. One of the important messages we plan to convey to the general public is that well-being during pregnancy and during the perinatal period has extremely important implications for healthy living throughout the life course.

SCHOOLS: our research project is particularly well suited for presentation in schools. This is because vision is intuitive and pupils can directly relate to it. We will endeavour to provide interactions with pupils. This will impact not only on their general understanding of science and their awareness of the importance of early experience throughout the lifespan, but also possibly on their future choices to enter a scientific discipline when building up their career.

Publications

10 25 50
 
Title new method for cell classification in the retina 
Description This new method enables classification of retinal ganglion cells into different functional classes without having to use very specific visual stimuli. It is simply based on the fact that cells belonging to the same class will tend to fire with the same firing patterns (distances between spikes in spikes trains in pairs of cells). The method can easily be applied to different sensory modalities. 
Type Of Material Physiological assessment or outcome measure 
Year Produced 2018 
Provided To Others? Yes  
Impact It is a bit too early to say, it was published in December 2018 It is drawing a lot of interest (from looking at paper metrics on the Frontiers in Neuroscience website). 
 
Description APS recordings collaborations 
Organisation Italian Institute of Technology (Istituto Italiano di Tecnologia IIT)
Country Italy 
Sector Public 
PI Contribution we developed the use of the high density array system to record from the retina
Collaborator Contribution They contributed the technology, the high density system as well as some analytical softwares At INRIA they help us with the analysis of the very complex data generated with the system
Impact multidisciplinary: neuroscience, electronic and software engineering outcomes: one paper several conference proceedings new funding acquisition of the hardware at great discount 2 more papers in review process
Start Year 2012
 
Description APS recordings collaborations 
Organisation The National Institute for Research in Computer Science and Control (INRIA)
Country France 
Sector Public 
PI Contribution we developed the use of the high density array system to record from the retina
Collaborator Contribution They contributed the technology, the high density system as well as some analytical softwares At INRIA they help us with the analysis of the very complex data generated with the system
Impact multidisciplinary: neuroscience, electronic and software engineering outcomes: one paper several conference proceedings new funding acquisition of the hardware at great discount 2 more papers in review process
Start Year 2012
 
Description Collaboration with Bruno Cessac INRIA 
Organisation French Institute for Research in Computer Science and Automation
Country France 
Sector Private 
PI Contribution Generating electrophysiological data on retinal waves and retinal receptive fields
Collaborator Contribution Bruno Cessac is interested in understanding the mechanisms underlying the generation and propagation of retinal waves from a theoretical point of view and he will use data generated by this project to test some models
Impact none yet
Start Year 2017
 
Description Holmes lecture 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Schools
Results and Impact Holmes lecture on the retina to high school children
Year(s) Of Engagement Activity 2018