Competition is a driving force in topographic mapping

Reanalysis of EphA3 Knock-In Double Maps in Mouse Suggests That Stochasticity in Topographic Map Formation Acts at the Retina Rather than between Competing Mechanisms at the Colliculus

It has been suggested that stochasticity acts in the formation of topographically ordered maps in the visual system through the opposing chemoaffinity and neural activity forces acting on the innervating nerve fibers being held in an unstable equilibrium. Evidence comes from the Islet2-EphA3 knock-in mouse, in which ∼50% of the retinal ganglion cells, distributed across the retina, acquire the EphA3 receptor, thus having an enhanced density of EphA which specifies retinotopic order along the rostrocaudal (RC) axis of the colliculus. Sampling EphA3 knock-in maps in heterozygotes at different positions along the mediolateral (ML) extent of the colliculus had found single 1D maps [as in wild types (WTs)], double maps (as in homozygous knock-ins) or both single and double maps. We constructed full 2D maps from the same mouse dataset. We found either single maps or maps where the visual field projects rostrally, with a part-projection more caudally to form a double map, the extent and location of this duplication varying considerably. Contrary to previous analyses, there was no strict demarcation between heterozygous and homozygous maps. These maps were replicated in a computational model where, as the level of EphA3 was increased, there was a smooth transition from single to double maps. Our results suggest that the diversity in these retinotopic maps has its origin in a variability over the retina in the effective amount of EphA3, such as through variability in gene expression or the proportion of EphA3+ retinal ganglion cells, rather than the result of competing mechanisms acting at the colliculus.

Complex computation from developmental priors

Machine learning (ML) models have long overlooked innateness: how strong pressures for survival lead to the encoding of complex behaviors in the nascent wiring of a brain. Here, we derive a neurodevelopmental encoding of artificial neural networks that considers the weight matrix of a neural network to be emergent from well-studied rules of neuronal compatibility. Rather than updating the network's weights directly, we improve task fitness by updating the neurons' wiring rules, thereby mirroring evolutionary selection on brain development. We find that our model (1) provides sufficient representational power for high accuracy on ML benchmarks while also compressing parameter count, and (2) can act as a regularizer, selecting simple circuits that provide stable and adaptive performance on metalearning tasks. In summary, by introducing neurodevelopmental considerations into ML frameworks, we not only model the emergence of innate behaviors, but also define a discovery process for structures that promote complex computations.

Activity-dependent Organization of Topographic Neural Circuits

Sensory information in the brain is organized into spatial representations, including retinotopic, somatotopic, and tonotopic maps, as well as ocular dominance columns. The spatial representation of sensory inputs is thought to be a fundamental organizational principle that is important for information processing. Topographic maps are plastic throughout an animal's life, reflecting changes in development and aging of brain circuitry, changes in the periphery and sensory input, and changes in circuitry, for instance in response to experience and learning. Here, we review mechanisms underlying the role of activity in the development, stability and plasticity of topographic maps, focusing on recent work suggesting that the spatial information in the visual field, and the resulting spatiotemporal patterns of activity, provide instructive cues that organize visual projections.

Segregation of Multimodal Inputs Into Discrete Midbrain Compartments During an Early Critical Period

The lateral cortex of the inferior colliculus (LCIC) is a multimodal subdivision of the midbrain inferior colliculus (IC) that plays a key role in sensory integration. The LCIC is compartmentally-organized, exhibiting a series of discontinuous patches or modules surrounded by an extramodular matrix. In adult mice, somatosensory afferents target LCIC modular zones, while auditory afferents terminate throughout the encompassing matrix. Recently, we defined an early LCIC critical period (birth: postnatal day 0 to P12) based upon the concurrent emergence of its neurochemical compartments (modules: glutamic acid decarboxylase, GAD+; matrix: calretinin, CR+), matching Eph-ephrin guidance patterns, and specificity of auditory inputs for its matrix. Currently lacking are analogous experiments that address somatosensory afferent shaping and the construction of discrete LCIC multisensory maps. Combining living slice tract-tracing and immunocytochemical approaches in a developmental series of GAD67-GFP knock-in mice, the present study characterizes: (1) the targeting of somatosensory terminals for emerging LCIC modular fields; and (2) the relative separation of somatosensory and auditory inputs over the course of its established critical period. Results indicate a similar time course and progression of LCIC projection shaping for both somatosensory (corticocollicular) and auditory (intracollicular) inputs. While somewhat sparse and intermingling at birth, modality-specific projection patterns soon emerge (P4–P8), coincident with peak guidance expression and the appearance of LCIC compartments. By P12, an adult-like arrangement is in place, with fully segregated multimodal afferent arrays. Quantitative measures confirm increasingly distinct input maps, exhibiting less projection overlap with age. Potential mechanisms whereby multisensory LCIC afferent systems recognize and interface with its emerging modular-matrix framework are discussed.

NMDA Receptor Expression by Retinal Ganglion Cells Is Not Required for Retinofugal Map Formation nor Eye-Specific Segregation in the Mouse

Retinal ganglion cells (RGCs) project topographically to the superior colliculus (SC) and dorsal lateral geniculate nucleus (dLGN). Spontaneous activity plays a critical role in retinotopic mapping in both regions; however, the molecular mechanisms underlying activity-dependent refinement remain unclear. Previous pharmacologic studies implicate NMDA receptors (NMDARs) in the establishment of retinotopy. In other brain regions, NMDARs are expressed on both the presynaptic and postsynaptic side of the synapse, and recent work suggests that presynaptic and postsynaptic NMDARs play distinct roles in retinotectal developmental dynamics. To directly test the role of NMDARs expressed by RGCs in retinofugal map formation, we took a conditional genetic knock-out approach to delete the obligate GluN1 subunit of NMDARs in RGCs. Here, we demonstrate reduced GluN1 expression in the retina of Chrnb3-Cre;GluN1^flox/flox (pre-cKO) mice without altered expression in the SC. Anatomical tracing experiments revealed no significant changes in termination zone size in the SC and dLGN of pre-cKO mice, suggesting NMDAR function in RGCs is not an absolute requirement for topographic refinement. Further, we observed no change in the eye-specific organization of retinal inputs to the SC nor dLGN. To verify that NMDA induces activity in RGC terminals, we restricted GCaMP5 expression to RGCs and confirmed induction of calcium transients in RGC terminals. Together, these findings demonstrate that NMDARs expressed by RGCs are not required for retinofugal topographic map formation nor eye-specific segregation in the mouse.

Atoh7-independent specification of retinal ganglion cell identity

Retinal ganglion cells (RGCs) relay visual information from the eye to the brain. RGCs are the first cell type generated during retinal neurogenesis. Loss of function of the transcription factor Atoh7, expressed in multipotent early neurogenic retinal progenitors leads to a selective and essentially complete loss of RGCs. Therefore, Atoh7 is considered essential for conferring competence on progenitors to generate RGCs. Despite the importance of Atoh7 in RGC specification, we find that inhibiting apoptosis in Atoh7-deficient mice by loss of function of Bax only modestly reduces RGC numbers. Single-cell RNA sequencing of Atoh7;Bax-deficient retinas shows that RGC differentiation is delayed but that the gene expression profile of RGC precursors is grossly normal. Atoh7;Bax-deficient RGCs eventually mature, fire action potentials, and incorporate into retinal circuitry but exhibit severe axonal guidance defects. This study reveals an essential role for Atoh7 in RGC survival and demonstrates Atoh7-dependent and Atoh7-independent mechanisms for RGC specification.

Wiring subcortical image-forming centers: Topography, laminar targeting, and map alignment

Efficient sensory processing is a complex and important function for species survival. As such, sensory circuits are highly organized to facilitate rapid detection of salient stimuli and initiate motor responses. For decades, the retina's projections to image-forming centers have served as useful models to elucidate the mechanisms by which such exquisite circuitry is wired. In this chapter, we review the roles of molecular cues, neuronal activity, and axon-axon competition in the development of topographically ordered retinal ganglion cell (RGC) projections to the superior colliculus (SC) and dorsal lateral geniculate nucleus (dLGN). Further, we discuss our current state of understanding regarding the laminar-specific targeting of subclasses of RGCs in the SC and its homolog, the optic tectum (OT). Finally, we cover recent studies examining the alignment of projections from primary visual cortex with RGCs that monitor the same region of space in the SC.

New insights on the modeling of the molecular mechanisms underlying neural maps alignment in the midbrain

We previously identified and modeled a principle of visual map alignment in the midbrain involving the mapping of the retinal projections and concurrent transposition of retinal guidance cues into the superior colliculus providing positional information for the organization of cortical V1 projections onto the retinal map (Savier et al., 2017). This principle relies on mechanisms involving Epha/Efna signaling, correlated neuronal activity and axon competition. Here, using the 3-step map alignment computational model, we predict and validate in vivo the visual mapping defects in a well-characterized mouse model. Our results challenge previous hypotheses and provide an alternative, although complementary, explanation for the phenotype observed. In addition, we propose a new quantification method to assess the degree of alignment and organization between maps, allowing inter-model comparisons. This work generalizes the validity and robustness of the 3-step map alignment algorithm as a predictive tool and confirms the basic mechanisms of visual map organization.

Trans-Axonal Signaling in Neural Circuit Wiring

The development of neural circuits is a complex process that relies on the proper navigation of axons through their environment to their appropriate targets. While axon–environment and axon–target interactions have long been known as essential for circuit formation, communication between axons themselves has only more recently emerged as another crucial mechanism. Trans-axonal signaling governs many axonal behaviors, including fasciculation for proper guidance to targets, defasciculation for pathfinding at important choice points, repulsion along and within tracts for pre-target sorting and target selection, repulsion at the target for precise synaptic connectivity, and potentially selective degeneration for circuit refinement. This review outlines the recent advances in identifying the molecular mechanisms of trans-axonal signaling and discusses the role of axon–axon interactions during the different steps of neural circuit formation.

Molecular complexity of visual mapping: a challenge for regenerating therapy

Investigating the cellular and molecular mechanisms involved in the development of topographically ordered connections in the central nervous system constitutes an important issue in neurobiology because these connections are the base of the central nervous system normal function. The dominant model to study the development of topographic maps is the projection from the retinal ganglion cells to the optic tectum/colliculus. The expression pattern of Eph/ephrin system in opposing gradients both in the retina and the tectum, labels the local addresses on the target and gives specific sensitivities to growth cones according to their topographic origin in the retina. The rigid precision of normal retinotopic mapping has prompted the chemoaffinity hypothesis, positing axonal targeting to be based on fixed biochemical affinities between fibers and targets. However, several lines of evidence have shown that the mapping can adjust to experimentally modified targets with flexibility, demonstrating the robustness of the guidance process. Here we discuss the complex ways the Ephs and ephrins interact allowing to understand how the retinotectal mapping is a precise but also a flexible process.

Network cloning using DNA barcodes

The connections between neurons determine the computations performed by a neural network. Connections can be considered a “summary” of the statistical structure of the experience—data—on which the network was trained. Here, we propose a method for how neuronal network connectivity can be copied or “cloned” from one network to another. Our method relies on the use of DNA barcodes—short DNA sequences that allow tagging individual neurons with unique labels. In our study, we prove theorems that show that such a transfer of network connectivity is theoretically possible.

The connections between neurons determine the computations performed by both artificial and biological neural networks. Recently, we have proposed SYNSeq, a method for converting the connectivity of a biological network into a form that can exploit the tremendous efficiencies of high-throughput DNA sequencing. In SYNSeq, each neuron is tagged with a random sequence of DNA—a “barcode”—and synapses are represented as barcode pairs. SYNSeq addresses the analysis problem, reducing a network into a suspension of barcode pairs. Here, we formulate a complementary synthesis problem: How can the suspension of barcode pairs be used to “clone” or copy the network back into an uninitialized tabula rasa network? Although this synthesis problem might be expected to be computationally intractable, we find that, surprisingly, this problem can be solved efficiently, using only neuron-local information. We present the “one-barcode–one-cell” (OBOC) algorithm, which forces all barcodes of a given sequence to coalesce into the same neuron, and show that it converges in a number of steps that is a power law of the network size. Rapid and reliable network cloning with single-synapse precision is thus theoretically possible.

A stochastic framework to model axon interactions within growing neuronal populations

The confined and crowded environment of developing brains imposes spatial constraints on neuronal cells that have evolved individual and collective strategies to optimize their growth. These include organizing neurons into populations extending their axons to common target territories. How individual axons interact with each other within such populations to optimize innervation is currently unclear and difficult to analyze experimentally in vivo. Here, we developed a stochastic model of 3D axon growth that takes into account spatial environmental constraints, physical interactions between neighboring axons, and branch formation. This general, predictive and robust model, when fed with parameters estimated on real neurons from the Drosophila brain, enabled the study of the mechanistic principles underlying the growth of axonal populations. First, it provided a novel explanation for the diversity of growth and branching patterns observed in vivo within populations of genetically identical neurons. Second, it uncovered that axon branching could be a strategy optimizing the overall growth of axons competing with others in contexts of high axonal density. The flexibility of this framework will make it possible to investigate the rules underlying axon growth and regeneration in the context of various neuronal populations.

Understanding how neuronal cells establish complex circuits with specific functions within a developing brain is a major current challenge. Over the last past years, enormous progress has been done to precisely resolve brain anatomy and to dissect the mechanisms controlling the establishment of precise neuronal networks. However, due to the extreme complexity of the brain, it is still experimentally difficult to investigate in vivo how neurons interact with each other and with their physical environments to innervate target territories during development. Here, we have developed a framework that integrates a dynamic 3D mathematical model of single axonal growth with parameters estimated from neurons grown in vivo and simulations of entire populations of growing axons. The emergent properties of our model enable the study of the mechanistic principles underlying the growth of axonal population in developing brains. Specifically, our results highlight the impact of mechanical interactions on both individual and collective axon growth, and uncover how branching regulate this process.

Theoretical Models of Neural Development

Constructing a functioning nervous system requires the precise orchestration of a vast array of mechanical, molecular, and neural-activity-dependent cues. Theoretical models can play a vital role in helping to frame quantitative issues, reveal mathematical commonalities between apparently diverse systems, identify what is and what is not possible in principle, and test the abilities of specific mechanisms to explain the data. This review focuses on the progress that has been made over the last decade in our theoretical understanding of neural development.

Visual subcircuit-specific dysfunction and input-specific mispatterning in the superior colliculus of fragile X mice

BACKGROUND

Sensory processing deficits are frequently co-morbid with neurodevelopmental disorders. For example, patients with fragile X syndrome (FXS), caused by a silencing of the FMR1 gene, exhibit impairments in visual function specific to the dorsal system, which processes motion information. However, the developmental and circuit mechanisms underlying this deficit remain unclear. Recently, the superior colliculus (SC), a midbrain structure regulating head and eye movements, has emerged as a model for dissecting visual circuit development and function. Previous studies have demonstrated a critical role for activity-dependent processes in the development of visual circuitry in the SC. Based on the known role of the FMR1 gene product in activity-dependent synaptic plasticity, we explored the function and organization of visual circuits in the SC of a mouse model of FXS (Fmr1^-/y).

METHODS

We utilized in vivo extracellular electrophysiology in combination with computer-controlled visual stimuli to determine the receptive field properties of visual neurons in the SC of control and Fmr1^-/y mice. In addition, we utilized anatomical tracing methods to assess the organization of visual inputs to the SC and along the retinogeniculocortical pathway.

RESULTS

Receptive fields of visual neurons in the SC of Fmr1^-/y mice were significantly larger than those found in control animals, though their shape and structure were unaffected. Further, selectivity for direction of movement was decreased, while selectivity to axis of movement was unchanged. Interestingly, axis-selective (AS) neurons exhibited a specific hyperexcitability in comparison to AS neurons in control SC and to direction-selective (DS) neurons in both control and Fmr1^-/y SC. Anatomical tracings revealed that retinocollicular, retinogeniculate, and geniculocortical projections were normally organized in the absence of Fmr1. However, projections from primary visual cortex (V1) to the SC were poorly refined.

CONCLUSIONS

Fmr1 is required for the proper development of visual circuit organization and function in the SC. We find that visual dysfunction is heterogeneously manifested in a subcircuit-specific manner in Fmr1^-/y mice, consistent with previous studies in human FXS patients. Further, we show a specific alteration of inputs to the SC from V1, but not the retina. Together, these data suggest that Fmr1 may function in distinct ways during the development of different visual subcircuits.

Developmental interactions between thalamus and cortex: a true love reciprocal story

The developmental programs that control the specification of cortical and thalamic territories are maintained largely as independent processes. However, bulk of evidence demonstrates the requirement of the reciprocal interactions between cortical and thalamic neurons as key for the correct development of functional thalamocortical circuits. This reciprocal loop of connections is essential for sensory processing as well as for the execution of complex sensory-motor tasks. Here, we review recent advances in our understanding of how mutual collaborations between both brain regions define area patterning and cell differentiation in the thalamus and cortex.

Teneurin-3 controls topographic circuit assembly in the hippocampus

Brain functions rely on specific patterns of connectivity. Teneurins are evolutionarily conserved transmembrane proteins that instruct synaptic partner matching in Drosophila and are required for vertebrate visual system development. The roles of vertebrate teneurins in connectivity beyond the visual system remain largely unknown and their mechanisms of action have not been demonstrated. Here we show that mouse teneurin-3 is expressed in multiple topographically interconnected areas of the hippocampal region, including proximal CA1, distal subiculum, and medial entorhinal cortex. Viral-genetic analyses reveal that teneurin-3 is required in both CA1 and subicular neurons for the precise targeting of proximal CA1 axons to distal subiculum. Furthermore, teneurin-3 promotes homophilic adhesion in vitro in a splicing isoform-dependent manner. These findings demonstrate striking genetic heterogeneity across multiple hippocampal areas and suggest that teneurin-3 may orchestrate the assembly of a complex distributed circuit in the mammalian brain via matching expression and homophilic attraction.

Novel Models of Visual Topographic Map Alignment in the Superior Colliculus

The establishment of precise neuronal connectivity during development is critical for sensing the external environment and informing appropriate behavioral responses. In the visual system, many connections are organized topographically, which preserves the spatial order of the visual scene. The superior colliculus (SC) is a midbrain nucleus that integrates visual inputs from the retina and primary visual cortex (V1) to regulate goal-directed eye movements. In the SC, topographically organized inputs from the retina and V1 must be aligned to facilitate integration. Previously, we showed that retinal input instructs the alignment of V1 inputs in the SC in a manner dependent on spontaneous neuronal activity; however, the mechanism of activity-dependent instruction remains unclear. To begin to address this gap, we developed two novel computational models of visual map alignment in the SC that incorporate distinct activity-dependent components. First, a Correlational Model assumes that V1 inputs achieve alignment with established retinal inputs through simple correlative firing mechanisms. A second Integrational Model assumes that V1 inputs contribute to the firing of SC neurons during alignment. Both models accurately replicate in vivo findings in wild type, transgenic and combination mutant mouse models, suggesting either activity-dependent mechanism is plausible. In silico experiments reveal distinct behaviors in response to weakening retinal drive, providing insight into the nature of the system governing map alignment depending on the activity-dependent strategy utilized. Overall, we describe novel computational frameworks of visual map alignment that accurately model many aspects of the in vivo process and propose experiments to test them.

Quantitative Analysis of Axonal Branch Dynamics in the Developing Nervous System

Branching is an important mechanism by which axons navigate to their targets during neural development. For instance, in the developing zebrafish retinotectal system, selective branching plays a critical role during both initial pathfinding and subsequent arborisation once the target zone has been reached. Here we show how quantitative methods can help extract new information from time-lapse imaging about the nature of the underlying branch dynamics. First, we introduce Dynamic Time Warping to this domain as a method for automatically matching branches between frames, replacing the effort required for manual matching. Second, we model branch dynamics as a birth-death process, i.e. a special case of a continuous-time Markov process. This reveals that the birth rate for branches from zebrafish retinotectal axons, as they navigate across the tectum, increased over time. We observed no significant change in the death rate for branches over this time period. However, blocking neuronal activity with TTX slightly increased the death rate, without a detectable change in the birth rate. Third, we show how the extraction of these rates allows computational simulations of branch dynamics whose statistics closely match the data. Together these results reveal new aspects of the biology of retinotectal pathfinding, and introduce computational techniques which are applicable to the study of axon branching more generally.

The complex morphologies of neurons present challenges for analysis. Large data sets can be gathered, but extracting meaningful data from the hundreds of branches from one axon over a few hundred time points can be difficult. One problem in particular is matching a single unique branch through several images, when the branches can extend, retract, or be removed entirely. In addition, if the imaging is done in vivo, the environment itself can grow and shift. Here we introduce Dynamic Time Warping (DTW) analysis to follow the complex structures of neurons through time. DTW identifies individual branches and therefore allows the determination of branch lifetimes. Using this approach we find that for retinal ganglion cell axons, the branch birth rate increases over time as axons navigate to their targets, and that blocking neural activity slightly increases the branch death rate without impacting the birth rate. From the estimated birth and death rate parameters we create simulations based on a continuous-time Markov chain process. These tools expand the techniques available to study the development of neuronal structures and provide more information from large time-lapse imaging datasets.

Stochastic Interaction between Neural Activity and Molecular Cues in the Formation of Topographic Maps

Topographic maps in visual processing areas maintain the spatial order of the visual world. Molecular cues and neuronal activity both play critical roles in map formation, but their interaction remains unclear. Here, we demonstrate that when molecular- and activity-dependent cues are rendered nearly equal in force, they drive topographic mapping stochastically. The functional and anatomical representation of azimuth in the superior colliculus of heterozygous Islet2-EphA3 knockin (Isl2(EphA3/+)) mice is variable: maps may be single, duplicated, or a combination of the two. This heterogeneity is not due to genetic differences, since map organizations in individual mutant animals often differ between colliculi. Disruption of spontaneous waves of retinal activity resulted in uniform map organization in Isl2(EphA3/+) mice, demonstrating that correlated spontaneous activity is required for map heterogeneity. Computational modeling replicates this heterogeneity, revealing that molecular- and activity-dependent forces interact simultaneously and stochastically during topographic map formation.

Developmental remodeling of relay cells in the dorsal lateral geniculate nucleus in the absence of retinal input

Background

The dorsal lateral geniculate nucleus (dLGN) of the mouse has been an important experimental model for understanding thalamic circuit development. The developmental remodeling of retinal projections has been the primary focus, however much less is known about the maturation of their synaptic targets, the relay cells of the dLGN. Here we examined the growth and maturation of relay cells during the first few weeks of life and addressed whether early retinal innervation affects their development. To accomplish this we utilized the math5 null (math5^−/−) mouse, a mutant lacking retinal ganglion cells and central projections.

Results

The absence of retinogeniculate axon innervation led to an overall shrinkage of dLGN and disrupted the pattern of dendritic growth among developing relay cells. 3-D reconstructions of biocytin filled neurons from math5^−/− mice showed that in the absence of retinal input relay cells undergo a period of exuberant dendritic growth and branching, followed by branch elimination and an overall attenuation in dendritic field size. However, math5^−/− relay cells retained a sufficient degree of complexity and class specificity, as well as their basic membrane properties and spike firing characteristics.

Conclusions

Retinal innervation plays an important trophic role in dLGN development. Additional support perhaps arising from non-retinal innervation and signaling is likely to contribute to the stabilization of their dendritic form and function.

Spatial pattern of spontaneous retinal waves instructs retinotopic map refinement more than activity frequency

Spontaneous activity during early development is necessary for the formation of precise neural connections, but it remains uncertain whether activity plays an instructive or permissive role in brain wiring. In the visual system, retinal ganglion cell (RGC) projections to the brain form two prominent sensory maps, one reflecting eye of origin and the other retinotopic location. Recent studies provide compelling evidence supporting an instructive role for spontaneous retinal activity in the development of eye-specific projections, but evidence for a similarly instructive role in the development of retinotopy is more equivocal. Here, we report on experiments in which we knocked down the expression of β2-containing nicotinic acetylcholine receptors (β2-nAChRs) specifically in the retina through a Cre-loxP recombination strategy. Overall levels of spontaneous retinal activity in retina-specific β2-nAChR mutant mice (Rx-β2cKO), examined in vitro and in vivo, were reduced to a degree comparable to that observed in whole animal β2-nAChR mouse mutants (β2KO). However, many residual spontaneous waves in Rx-β2cKO mice displayed local propagating features with strong correlations between nearby but not distant RGCs typical of waves observed in wild-type (WT) but not β2KO mice. We further observed that eye-specific segregation was disrupted in Rx-β2cKO mice, but retinotopy was spared in a competition-dependent manner. These results suggest that propagating patterns of spontaneous retinal waves are essential for normal development of the retinotopic map, even while overall activity levels are significantly reduced, and support an instructive role for spontaneous retinal activity in both eye-specific segregation and retinotopic refinement.

Ephrin-As are required for the topographic mapping but not laminar choice of physiologically distinct RGC types

In the retinocollicular projection, the axons from functionally distinct retinal ganglion cell (RGC) types form synapses in a stereotypical manner along the superficial to deep axis of the superior colliculus (SC). Each lamina contains an orderly topographic map of the visual scene but different laminae receive inputs from distinct sets of RGCs, and inputs to each lamina are aligned with the others to integrate parallel streams of visual information. To determine the relationship between laminar organization and topography of physiologically defined RGC types, we used genetic and anatomical axon tracing techniques in wild type and ephrin-A mutant mice. We find that adjacent RGCs of the same physiological type can send axons to both ectopic and normal topographic locations, supporting a penetrance model for ephrin-A independent mapping cues. While the overall laminar organization in the SC is unaffected in ephrin-A2/A5 double mutant mice, analysis of the laminar locations of ectopic terminations shows that the topographic maps of different RGC types are misaligned. These data lend support to the hypothesis that the retinocollicular projection is a superimposition of a number of individual two-dimensional topographic maps that originate from specific types of RGCs, require ephrin-A signaling, and form independently of the other maps.

Quantitative assessment of computational models for retinotopic map formation

Molecular and activity-based cues acting together are thought to guide retinal axons to their terminal sites in vertebrate optic tectum or superior colliculus (SC) to form an ordered map of connections. The details of mechanisms involved, and the degree to which they might interact, are still not well understood. We have developed a framework within which existing computational models can be assessed in an unbiased and quantitative manner against a set of experimental data curated from the mouse retinocollicular system. Our framework facilitates comparison between models, testing new models against known phenotypes and simulating new phenotypes in existing models. We have used this framework to assess four representative models that combine Eph/ephrin gradients and/or activity-based mechanisms and competition. Two of the models were updated from their original form to fit into our framework. The models were tested against five different phenotypes: wild type, Isl2-EphA3(ki/ki), Isl2-EphA3(ki/+), ephrin-A2,A3,A5 triple knock-out (TKO), and Math5(-/-) (Atoh7). Two models successfully reproduced the extent of the Math5(-/-) anteromedial projection, but only one of those could account for the collapse point in Isl2-EphA3(ki/+). The models needed a weak anteroposterior gradient in the SC to reproduce the residual order in the ephrin-A2,A3,A5 TKO phenotype, suggesting either an incomplete knock-out or the presence of another guidance molecule. Our article demonstrates the importance of testing retinotopic models against as full a range of phenotypes as possible, and we have made available MATLAB software, we wrote to facilitate this process.

Optic flow instructs retinotopic map formation through a spatial to temporal to spatial transformation of visual information

Retinotopic maps are plastic in response to changes in sensory input; however, the experience-dependent instructive cues that organize retinotopy are unclear. In animals with forward-directed locomotion, the predominant anterior to posterior optic flow activates retinal ganglion cells in a stereotyped temporal to nasal sequence. Here we imaged retinotectal axon arbor location and structural plasticity to assess map refinement in vivo while exposing Xenopus tadpoles to visual stimuli. We show that the temporal sequence of retinal activity driven by natural optic flow organizes retinotopy by regulating axon arbor branch dynamics, whereas the opposite sequence of retinal activity prevents map refinement. Our study demonstrates that a spatial to temporal to spatial transformation of visual information controls experience-dependent topographic map plasticity. This organizational principle is likely to apply to other sensory modalities and projections in the brain.

Axon-Schwann cell interactions during peripheral nerve regeneration in zebrafish larvae

Background

Peripheral nerve injuries can severely affect the way that animals perceive signals from the surrounding environment. While damage to peripheral axons generally has a better outcome than injuries to central nervous system axons, it is currently unknown how neurons re-establish their target innervations to recover function after injury, and how accessory cells contribute to this task. Here we use a simple technique to create reproducible and localized injury in the posterior lateral line (pLL) nerve of zebrafish and follow the fate of both neurons and Schwann cells.

Results

Using pLL single axon labeling by transient transgene expression, as well as transplantation of glial precursor cells in zebrafish larvae, we individualize different components in this system and characterize their cellular behaviors during the regenerative process. Neurectomy is followed by loss of Schwann cell differentiation markers that is reverted after nerve regrowth. We show that reinnervation of lateral line hair cells in neuromasts during pLL nerve regeneration is a highly dynamic process with promiscuous yet non-random target recognition. Furthermore, Schwann cells are required for directional extension and fasciculation of the regenerating nerve. We provide evidence that these cells and regrowing axons are mutually dependant during early stages of nerve regeneration in the pLL. The role of ErbB signaling in this context is also explored.

Conclusion

The accessibility of the pLL nerve and the availability of transgenic lines that label this structure and their synaptic targets provides an outstanding in vivo model to study the different events associated with axonal extension, target reinnervation, and the complex cellular interactions between glial cells and injured axons during nerve regeneration.

Computational and mathematical methods for morphogenetic gradient analysis, boundary formation and axonal targeting

Morphogenesis and axonal targeting are key processes during development that depend on complex interactions at molecular, cellular and tissue level. Mathematical modeling is essential to bridge this multi-scale gap in order to understand how the emergence of large structures is controlled at molecular level by interactions between various signaling pathways. We summarize mathematical modeling and computational methods for time evolution and precision of morphogenetic gradient formation. We discuss tissue patterning and the formation of borders between regions labeled by different morphogens. Finally, we review models and algorithms that reveal the interplay between morphogenetic gradients and patterned activity for axonal pathfinding and the generation of the retinotopic map in the visual system.

Chemoaffinity in topographic mapping revisited--is it more about fiber-fiber than fiber-target interactions?

Axonal projections between two populations of neurons, which preserve neighborhood relationships, are called topographic. They are ubiquitous in the brain. The development of the retinotectal projection, mapping the retinal output onto the roof of the midbrain, has been studied for decades as a model system. The rigid precision of normal retinotopic mapping has prompted the chemoaffinity hypothesis, positing axonal targeting to be based on fixed biochemical affinities between fibers and targets. In addition, however, abundant evidence has been gathered mainly in the 1970s and 80s that the mapping can adjust to variegated targets with stunning flexibility demonstrating the extraordinary robustness of the guidance process. The identification of ephrins and Eph-receptors as the underlying molecular cues has mostly been interpreted as supporting the fiber-target chemoaffinity hypothesis, while the evidence on mapping robustness has largely been neglected. By having a fresh look on the old data, we expound that they indicate, in addition to fiber-target chemoaffinity, the existence of a second autonomous guidance influence, which we call fiber-fiber chemoaffinity. Classical in vitro observations suggest both influences be composed of opposing monofunctional guidance activities. Based on the molecular evidence, we propose that those might be ephrin/Eph forward and reverse signaling, not only in fiber-target but also in fiber-fiber interactions. In fact, computational models based on this assumption can reconcile the seemingly conflicting findings on rigid and flexible topographic mapping. Supporting the suggested parsimonious and powerful mechanism, they contribute to an understanding of the evolutionary success of robust topographic mass wiring of axons.

Ephrin signalling in the developing nervous system

Ephrin ligands and their Eph receptors hold our attention since their link to axon guidance almost twenty years ago. Since then, they have been shown to be critical for short distance cell-cell interactions in the nervous system. The interest in their function has not abated, leading to ever-more sophisticated studies generating as many surprising answers about their function as new questions. We discuss recent insights into their functions in the developing nervous system, including neuronal progenitor sorting, stochastic cell migration, guidance of neuronal growth cones, topographic map formation, as well as synaptic plasticity.

Retinal overexpression of Ten-m3 alters ipsilateral retinogeniculate projections in the wallaby (Macropus eugenii)

The dorsal lateral geniculate nucleus (dLGN) contains a retinotopic map where input from the two eyes map in register to provide a substrate for binocular vision. Ten-m3, a transmembrane protein, mediates homophilic interactions and has been implicated in the patterning of ipsilateral visual projections. Ease of access to early developmental stages in a marsupial wallaby has been used to manipulate levels of Ten-m3 during the development of retinogeniculate projections. In situ hybridisation showed a high dorsomedial to low ventrolateral gradient of Ten-m3 in the developing dLGN, matching retinotopically with the previously reported high ventral to low dorsal retinal gradient. Overexpression of Ten-m3 in ventronasal but not dorsonasal retina resulted in an extension of ipsilateral projections beyond the normal binocular zone. These results demonstrate that Ten-m3 influences ipsilateral projections and support a role for it in binocular mapping.

The molecular basis for the development of neural maps

Neural development leads to the establishment of precise connectivity in the nervous system. By contrasting the information capacities of cortical connectivity and the genome, we suggest that simplifying rules are necessary in order to create cortical connections from the limited set of instructions contained in the genome. One of these rules may be employed by the visual system, where connections are formed on the basis of the interplay of molecular gradients and activity-dependent synaptic plasticity. We show how a simple model that accounts for such interplay can create both neural topographic maps and more complex patterns of ocular dominance, that is, the segregated binary mixture of projections from two eyes converging in the same visual area. With regard to the ocular dominance patterns, we show that pattern orientation may be instructed by the direction of the gradients of molecular labels. We also show that the periodicity of ocular dominance patterns may result from the interplay of the effects of molecular gradients and correlated neural activity. Overall, we propose that simple mechanisms can account for the formation of apparently complex features of neuronal connections.

Dendritic and axonal targeting patterns of a genetically-specified class of retinal ganglion cells that participate in image-forming circuits

Background

There are numerous functional types of retinal ganglion cells (RGCs), each participating in circuits that encode a specific aspect of the visual scene. This functional specificity is derived from distinct RGC morphologies and selective synapse formation with other retinal cell types; yet, how these properties are established during development remains unclear. Islet2 (Isl2) is a LIM-homeodomain transcription factor expressed in the developing retina, including approximately 40% of all RGCs, and has previously been implicated in the subtype specification of spinal motor neurons. Based on this, we hypothesized that Isl2⁺ RGCs represent a related subset that share a common function.

Results

We morphologically and molecularly characterized Isl2⁺ RGCs using a transgenic mouse line that expresses GFP in the cell bodies, dendrites and axons of Isl2⁺ cells (Isl2-GFP). Isl2-GFP RGCs have distinct morphologies and dendritic stratification patterns within the inner plexiform layer and project to selective visual nuclei. Targeted filling of individual cells reveals that the majority of Isl2-GFP RGCs have dendrites that are monostratified in layer S3 of the IPL, suggesting they are not ON-OFF direction-selective ganglion cells. Molecular analysis shows that most alpha-RGCs, indicated by expression of SMI-32, are also Isl2-GFP RGCs. Isl2-GFP RGCs project to most retino-recipient nuclei during early development, but specifically innervate the dorsal lateral geniculate nucleus and superior colliculus (SC) at eye opening. Finally, we show that the segregation of Isl2⁺ and Isl2^- RGC axons in the SC leads to the segregation of functional RGC types.

Conclusions

Taken together, these data suggest that Isl2⁺ RGCs comprise a distinct class and support a role for Isl2 as an important component of a transcription factor code specifying functional visual circuits. Furthermore, this study describes a novel genetically-labeled mouse line that will be a valuable resource in future investigations of the molecular mechanisms of visual circuit formation.

Restricted perinatal retinal degeneration induces retina reshaping and correlated structural rearrangement of the retinotopic map

The formation of the retinotopic map depends on the action of axon guidance molecules, activity-dependent mechanisms and axonal competition. However, little is known about the plasticity potential of the system and the effects on the remodelling of retinocollicular connections upon retinal insults. Here we create a mouse model in which retinal ganglion cells that project to anterior and posterior superior colliculus undergo cell death during topographic map formation. We show that the remaining retinal ganglion cells expand the targeted area in the superior colliculus and at the same time increase their spatial coverage in the retina in a correlated fashion. The resulting contralateral topographic map is overall maintained but less precise, while ipsilateral retinal ganglion cell axons are abnormally distributed in anterior and posterior superficial superior colliculus. These results suggest the presence of plastic mechanisms in the developing mammalian visual system to adjust retinal space and its target coverage and ensure a uniform map.

Neuron glia-related cell adhesion molecule (NrCAM) promotes topographic retinocollicular mapping

NrCAM (Neuron-glial related cell adhesion molecule), a member of the L1 family of cell adhesion molecules, reversibly binds ankyrin and regulates axon growth, but it has not been studied for a role in retinotopic mapping. During development of retino-collicular topography, NrCAM was expressed uniformly in retinal ganglion cells (RGCs) along both mediolateral and anteroposterior retinal axes, and was localized on RGC axons within the optic tract and superior colliculus (SC). Anterograde tracing of RGC axons in NrCAM null mutant mice at P10, when the map resembles its mature form, revealed laterally displaced ectopic termination zones (eTZs) of axons from the temporal retina, indicating defective mediolateral topography, which is governed by ephrinB/EphBs. Axon tracing at P2 revealed that interstitial branch orientation of ventral-temporal RGC axons in NrCAM null mice was compromised in the medial direction, likely accounting for displacement of eTZs. A similar retinocollicular targeting defect in EphB mutant mice suggested that NrCAM and EphB interact to regulate mediolateral retino-collicular targeting. We found that EphB2 tyrosine kinase but not an EphB2 kinase dead mutant, phosphorylated NrCAM at a conserved tyrosine residue in the FIGQY ankyrin binding motif, perturbing ankyrin recruitment in NrCAM transfected HEK293 cells. Furthermore, the phosphorylation of NrCAM at FIGQY in SC was decreased in EphB1/3 and EphB1/2/3 null mice compared to WT, while phospho-FIGQY of NrCAM in SC was increased in EphB2 constitutively active (F620D/F620D) mice. These results demonstrate that NrCAM contributes to mediolateral retinocollicular axon targeting by regulating RGC branch orientation through a likely mechanism in which ephrinB/EphB phosphorylates NrCAM to modulate linkage to the actin cytoskeleton.

Molecular guidance of retinotopic map development in the midbrain

Topographic maps are utilized in many sensory and motor systems to efficiently transfer information between brain regions. The retina's projection to the superior colliculus has served as a model for the identification of molecular cues and mechanistic strategies by which topographic maps are formed. Evidence from both in vitro and in vivo studies points to graded cell surface cues playing a central role, but support for axon-axon competition and selective degeneration have also been advanced recently. In combination with mathematical models, these studies suggest that topographic maps are established using a complex combination of strategies to ensure precise connectivity.

Retinocollicular mapping explained?

We review and comment on the recent model of Grimbert and Cang of the development of topographically ordered maps from the retina to the superior colliculus. This model posits a phase in which arbors are created in zones permitted by Eph and ephrin signaling, followed by a phase in which activity-dependent synaptic plasticity refines the map. We show that it is not possible to generate the arborization probability functions used in the simulations of Grimbert and Cang using gradients of Ephs and ephrins and the interaction mechanism that Grimbert and Cang propose in their results. Furthermore, the arborization probabilities we do generate are far less sharp than we imagine truly “permissive” ones would be. It remains to be seen if maps can be generated from the nonpermissive arborization probabilities generated from gradients.

What axons tell each other: axon-axon signaling in nerve and circuit assembly

A remarkable feature of nervous system development is the ability of axons emerging from newly formed neurons to traverse, by cellular scale, colossal distances to appropriate targets. The earliest axons achieve this in an essentially axon-free environment, but the vast majority of axons eventually grow along a scaffold of nerve tracts created by earlier extending axons. Signal exchange between sequentially or simultaneously extending axons may well represent the predominant mode of axonal navigation, but proportionally few efforts have so far been directed at deciphering the underlying mechanisms. This review intends to provide a conceptual update on the cellular and molecular principles driving axon-axon interactions, with emphasis on those contributing to the fidelity of axonal navigation, sorting and connectivity during nerve and circuit assembly.

On the Importance of Countergradients for the Development of Retinotopy: Insights from a Generalised Gierer Model

During the development of the topographic map from vertebrate retina to superior colliculus (SC), EphA receptors are expressed in a gradient along the nasotemporal retinal axis. Their ligands, ephrin-As, are expressed in a gradient along the rostrocaudal axis of the SC. Countergradients of ephrin-As in the retina and EphAs in the SC are also expressed. Disruption of any of these gradients leads to mapping errors. Gierer's (1981) model, which uses well-matched pairs of gradients and countergradients to establish the mapping, can account for the formation of wild type maps, but not the double maps found in EphA knock-in experiments. I show that these maps can be explained by models, such as Gierer's (1983), which have gradients and no countergradients, together with a powerful compensatory mechanism that helps to distribute connections evenly over the target region. However, this type of model cannot explain mapping errors found when the countergradients are knocked out partially. I examine the relative importance of countergradients as against compensatory mechanisms by generalising Gierer's (1983) model so that the strength of compensation is adjustable. Either matching gradients and countergradients alone or poorly matching gradients and countergradients together with a strong compensatory mechanism are sufficient to establish an ordered mapping. With a weaker compensatory mechanism, gradients without countergradients lead to a poorer map, but the addition of countergradients improves the mapping. This model produces the double maps in simulated EphA knock-in experiments and a map consistent with the Math5 knock-out phenotype. Simulations of a set of phenotypes from the literature substantiate the finding that countergradients and compensation can be traded off against each other to give similar maps. I conclude that a successful model of retinotopy should contain countergradients and some form of compensation mechanism, but not in the strong form put forward by Gierer.

Developmental mechanisms of topographic map formation and alignment

Brain connections are organized into topographic maps that are precisely aligned both within and across modalities. This alignment facilitates coherent integration of different categories of sensory inputs and allows for proper sensorimotor transformations. Topographic maps are established and aligned by multistep processes during development, including interactions of molecular guidance cues expressed in gradients; spontaneous activity-dependent axonal and dendritic remodeling; and sensory-evoked plasticity driven by experience. By focusing on the superior colliculus, a major site of topographic map alignment for different sensory modalities, this review summarizes current understanding of topographic map development in the mammalian visual system and highlights recent advances in map alignment studies. A major goal looking forward is to reveal the molecular and synaptic mechanisms underlying map alignment and to understand the physiological and behavioral consequences when these mechanisms are disrupted at various scales.

Sensory map transfer to the neocortex relies on pretarget ordering of thalamic axons

Sensory maps, such as the representation of mouse facial whiskers, are conveyed throughout the nervous system by topographic axonal projections that preserve neighboring relationships between adjacent neurons. In particular, the map transfer to the neocortex is ensured by thalamocortical axons (TCAs), whose terminals are topographically organized in response to intrinsic cortical signals. However, TCAs already show a topographic order early in development, as they navigate toward their target. Here, we show that this preordering of TCAs is required for the transfer of the whisker map to the neocortex. Using Ebf1 conditional inactivation that specifically perturbs the development of an intermediate target, the basal ganglia, we scrambled TCA topography en route to the neocortex without affecting the thalamus or neocortex. Notably, embryonic somatosensory TCAs were shifted toward the visual cortex and showed a substantial intermixing along their trajectory. Somatosensory TCAs rewired postnatally to reach the somatosensory cortex but failed to form a topographic anatomical or functional map. Our study reveals that sensory map transfer relies not only on positional information in the projecting and target structures but also on preordering of axons along their trajectory, thereby opening novel perspectives on brain wiring.

Developmental and architectural principles of the lateral-line neural map

The transmission and central representation of sensory cues through the accurate construction of neural maps is essential for animals to react to environmental stimuli. Structural diversity of sensorineural maps along a continuum between discrete- and continuous-map architectures can influence behavior. The mechanosensory lateral line of fishes and amphibians, for example, detects complex hydrodynamics occurring around the animal body. It triggers innate fast escape reactions but also modulates complex navigation behaviors that require constant knowledge about the environment. The aim of this article is to summarize recent work in the zebrafish that has shed light on the development and structure of the lateralis neural map, which is helping to understand how individual sensory modalities generate appropriate behavioral responses to the sensory context.

Regulation of ephrin-A expression in compressed retinocollicular maps

Retinotopic maps can undergo compression and expansion in response to changes in target size, but the mechanism underlying this compensatory process has remained a mystery. The discovery of ephrins as molecular mediators of Sperry's chemoaffinity process allows a mechanistic approach to this important issue. In Syrian hamsters, neonatal, partial (PT) ablation of posterior superior colliculus (SC) leads to compression of the retinotopic map, independent of neural activity. Graded, repulsive EphA receptor/ephrin-A ligand interactions direct the formation of the retinocollicular map, but whether ephrins might also be involved in map compression is unknown. To examine whether map compression might be directed by changes in the ephrin expression pattern, we compared ephrin-A2 and ephrin-A5 mRNA expression between normal SC and PT SC using in situ hybridization and quantitative real-time PCR. We found that ephrin-A ligand expression in the compressed maps was low anteriorly and high posteriorly, as in normal animals. Consistent with our hypothesis, the steepness of the ephrin gradient increased in the lesioned colliculi. Interestingly, overall levels of ephrin-A2 and -A5 expression declined immediately after neonatal target damage, perhaps promoting axon outgrowth. These data establish a correlation between changes in ephrin-A gradients and map compression, and suggest that ephrin-A expression gradients may be regulated by target size. This in turn could lead to compression of the retinocollicular map onto the reduced target. These findings have important implications for mechanisms of recovery from traumatic brain injury.

New model of retinocollicular mapping predicts the mechanisms of axonal competition and explains the role of reverse molecular signaling during development

Precise connections in the brain result from elaborate processes during development. In the visual system, axonal projections from retinal ganglion cells (RGCs) onto the superior colliculus form a precise retinotopic map. Studies have revealed that the development of retinocollicular maps involves three main factors: graded expression of molecular guidance cues such as EphAs and ephrin-As, activity-dependent processes driven by spontaneous activity in RGCs, and different forms of axonal competition. In this study, we developed a new, versatile model including these factors. We first modeled the selective arborization of RGC axons, mediated by EphA/ephrin-A signaling, without assuming that this initial process instructed the map's final topology. We also derived an integro-differential equation modeling a second, dynamic phase in which activity-dependent plasticity of axonal arbors combined with their competition for collicular resources can deeply remodel the topology of immature maps. Our model hence challenges the view that retinotopic maps are instructed by matching molecular gradients and then merely refined by activity-dependent processes. We reproduce fine features of retinotopic map development in wild-type and various transgenic mice, allowing a new understanding of the underlying mechanisms. Our model predicts that competition is not based on comparisons of axonal EphA receptor levels but rather relies on the optimization of collicular resources mediated by neurotrophic receptors such as p75(NTR). Our model finally clarifies the elusive role of reverse signaling between retinal ephrin-As and collicular EphAs by reproducing for the first time the phenotypes of two mouse genotypes in which this function is altered.

Pushing the envelope of retinal ganglion cell genesis: context dependent function of Math5 (Atoh7)

The basic-helix-loop helix factor Math5 (Atoh7) is required for retinal ganglion cell (RGC) development. However, only 10% of Math5-expressing cells adopt the RGC fate, and most become photoreceptors. In principle, Math5 may actively bias progenitors towards RGC fate or passively confer competence to respond to instructive factors. To distinguish these mechanisms, we misexpressed Math5 in a wide population of precursors using a Crx BAC or 2.4 kb promoter, and followed cell fates with Cre recombinase. In mice, the Crx cone-rod homeobox gene and Math5 are expressed shortly after cell cycle exit, in temporally distinct, but overlapping populations of neurogenic cells that give rise to 85% and 3% of the adult retina, respectively. The Crx>Math5 transgenes did not stimulate RGC fate or alter the timing of RGC births. Likewise, retroviral Math5 overexpression in retinal explants did not bias progenitors towards the RGC fate or induce cell cycle exit. The Crx>Math5 transgene did reduce the abundance of early-born (E15.5) photoreceptors two-fold, suggesting a limited cell fate shift. Nonetheless, retinal histology was grossly normal, despite widespread persistent Math5 expression. In an RGC-deficient (Math5 knockout) environment, Crx>Math5 partially rescued RGC and optic nerve development, but the temporal envelope of RGC births was not extended. The number of early-born RGCs (before E13) remained very low, and this was correlated with axon pathfinding defects and cell death. Together, these results suggest that Math5 is not sufficient to stimulate RGC fate. Our findings highlight the robust homeostatic mechanisms, and role of pioneering neurons in RGC development.

Math5 defines the ganglion cell competence state in a subpopulation of retinal progenitor cells exiting the cell cycle

The basic helix-loop-helix (bHLH) transcription factor Math5 (Atoh7) is transiently expressed during early retinal histogenesis and is necessary for retinal ganglion cell (RGC) development. Using nucleoside pulse-chase experiments and clonal analysis, we determined that progenitor cells activate Math5 during or after the terminal division, with progressively later onset as histogenesis proceeds. We have traced the lineage of Math5+ cells using mouse BAC transgenes that express Cre recombinase under strict regulatory control. Quantitative analysis showed that Math5+ progenitors express equivalent levels of Math5 and contribute to every major cell type in the adult retina, but are heavily skewed toward early fates. The Math5>Cre transgene labels 3% of cells in adult retina, including 55% of RGCs. Only 11% of Math5+ progenitors develop into RGCs; the majority become photoreceptors. The fate bias of the Math5 cohort, inferred from the ratio of cone and rod births, changes over time, in parallel with the remaining neurogenic population. Comparable results were obtained using Math5 mutant mice, except that ganglion cells were essentially absent, and late fates were overrepresented within the lineage. We identified Math5-independent RGC precursors in the earliest born (embryonic day 11) retinal cohort, but these precursors require Math5-expressing cells for differentiation. Math5 thus acts permissively to establish RGC competence within a subset of progenitors, but is not sufficient for fate specification. It does not autonomously promote or suppress the determination of non-RGC fates. These data are consistent with progressive and temporal restriction models for retinal neurogenesis, in which environmental factors influence the final histotypic choice.