Journal Article

Molecules, maps and synapse specificity

Colman, David R., Benson, Deanna L., Huntley, George W.

2001

Overview

Key Points Synaptogenesis is the culmination of a continuous process, which can be divided into the following stages: (1) axon guidance or pathfinding; (2) gross target recognition; (3) fine target recognition; and (4) elaboration of synaptic contacts onto appropriate cellular domains. Furthermore, synaptic connections are organized topographically, an essential anatomical substrate for orderly 'maps' of sensory surfaces, such as the retina. Sperry proposed that the topographically ordered distribution of synapses was established by “highly specific cytochemical affinities” between an axon and the environment through which it grows, and ultimately its target neuron. He proposed an orderly mapping of two or more standing gradients that are orthogonal to one another, so that an incoming axon is guided by signals encoding both latitude and longitude. Subsequent models have addressed the nature of standing gradients, and how a growth cone might sense and respond to the subtle differences in the molecular environment generated by such gradients. Haydon and Drapeau proposed two general modes of synapse specification. 'Selective' neurons send their neurites only to their appropriate target; 'promiscuous' neurons form synapses with a number of targets, and final specificity is achieved by pruning away the incorrect terminal sites in an activity-mediated process. Neuronal differentiation is the first step in synapse specification. Neurons, and the position they hold within a larger group, impart information. Group identification might be encoded, at least in part, by differential adhesion, and neighbour relationships within groups might be established by gap-junction-mediated communication, or by regulated patterns of calcium waves. The final topographic order of axons within a target might reflect an ordered distribution of axons within a fibre tract. However, retinal axon ordering alone does not seem to be sufficient for dorsoventral patterning in the optic tectum. In the dorsal thalamus, collections of neurons born contemporaneously parse into distinct nuclei. It is remarkable that targeting is precise from the earliest stages of innervation, because thalamic axons from different nuclei travel together through a similar environment, and are presented with an array of possible areal targets. Presynaptic assembly cannot be entirely nonspecific, or all potential partners brought into close proximity would form synapses with each other. Evidence indicates that a particular recognition threshold must be passed in order for synapse-initiation molecules to link. In vitro studies indicate that an interaction between β-neurexin and neuroligins can trigger synapse initiation. Several other molecules have been suggested to be involved in the early stages of synapse recognition/initiation, including EphB and Narp. Stabilizing a synapse is likely to require various molecules, but activity seems to be essential; strong evidence indicates that neurotrophins are involved, and recent work indicates that local synthesis of synaptic proteins might also be important. In Drosophila , homophilic binding between pre- and postsynaptically localized Fasciclin II is required to maintain a neuromuscular synapse, and members of the cadherin superfamily might have a similar role in vertebrates. Synaptogenesis should be viewed as an ongoing process that includes the modification and elimination of existing synapses and the generation of new synapses. Consistent with this, several guidance and recognition molecules continue to be expressed in adult nervous systems, and many have been implicated in the generation of synapse plasticity. A striking feature of the mature central nervous system is the precision of the synaptic circuitry. In contemplating the mature circuitry, it is impossible to imagine how more than 20 billion neurons in the human brain become precisely connected through trillions of synapses. Remarkably, much of the final wiring can be established in the absence of neural activity or experience; so the algorithms that allow precise connectivity must be encoded largely by the genetic programme. This programme, honed over nearly one billion years of evolution, generates networks with the flexibility to respond to a wide range of physiological challenges. There are several contemporary models of how synapse specificity is achieved, many of them proposed before the identification of guidance or recognition molecules. Here we review a selection of models as frameworks for defining the nature and complexity of synaptogenesis, and evaluate their validity in view of progress made in identifying the molecular underpinnings of axon guidance, targeting and synapse formation.

• Publisher: Nature Publishing Group UK,Nature Publishing Group
• Subject: Animal Genetics and Genomics / Animals / Behavioral Sciences / Biological Techniques / Biomedical and Life Sciences / Biomedicine / Body Patterning - physiology / Cell Communication - physiology / Cell Differentiation - physiology / Central Nervous System - embryology / Central Nervous System - growth & development / Central Nervous System - metabolism / Humans / Models, Biological / Nerve Growth Factors - metabolism / Nerve Net - embryology / Nerve Net - growth & development / Nerve Net - metabolism / Neurobiology / Neurons / Neurosciences / review-article / Synapses - metabolism / Synapses - ultrastructure / Target recognition / Topography