NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Alberts B, Johnson A, Lewis J, et al. Molecular Biology of the Cell. 4th edition. New York: Garland Science; 2002.

  • By agreement with the publisher, this book is accessible by the search feature, but cannot be browsed.
Cover of Molecular Biology of the Cell

Molecular Biology of the Cell. 4th edition.

Show details

Neural Development

Nerve cells, or neurons, are among the most ancient of all specialized animal cell types. Their structure is like that of no other class of cells, and the development of the nervous system poses problems that have no real parallel in other tissues. A neuron is extraordinary above all for its enormously extended shape, with a long axon and branching dendrites connecting it through synapses to other cells (Figure 21-88). The central challenge of neural development is to explain how the axons and dendrites grow out, find their right partners, and synapse with them selectively to create a functional network (Figure 21-89). The problem is formidable: the human brain contains more than 1011 neurons, each of which, on average, has to make connections with a thousand others, according to a regular and predictable wiring plan. The precision required is not so great as in a man-made computer, for the brain performs its computations in a different way and is more tolerant of vagaries in individual components; but the brain nevertheless outstrips all other biological structures in its organized complexity.

Figure 21-88. A typical neuron of a vertebrate.

Figure 21-88

A typical neuron of a vertebrate. The arrows indicate the direction in which signals are conveyed. The neuron shown is from the retina of a monkey. The longest and largest neurons in a human extend for about 1 million μm and have an axon diameter (more...)

Figure 21-89. The complex organization of nerve cell connections.

Figure 21-89

The complex organization of nerve cell connections. This drawing depicts a section through a small part of a mammalian brain—the olfactory bulb of a dog, stained by the Golgi technique. The black objects are neurons; the thin lines are axons and (more...)

The components of a typical nervous system—the various classes of neurons, glial cells, sensory cells, and muscles—originate in a number of widely separate locations in the embryo and are initially unconnected. Thus, in the first phase of neural development (Figure 21-90), the different parts develop according to their own local programs: neurons are born and assigned specific characters according to the place and time of their birth, under the control of inductive signals and gene regulatory mechanisms similar to those we have already discussed for other tissues of the body. The next phase involves a type of morphogenesis unique to the nervous system: axons and dendrites grow out along specific routes, setting up a provisional but orderly network of connections between the separate parts of the system. In the third and final phase, which continues into adult life, the connections are adjusted and refined through interactions among the far-flung components in a way that depends on the electrical signals that pass between them.

Figure 21-90. The three phases of neural development.

Figure 21-90

The three phases of neural development.

Neurons Are Assigned Different Characters According to the Time and Place Where They Are Born

Neurons are almost always produced in association with glial cells, which provide a supporting framework and create an enclosed, protected environment in which the neurons can perform their functions. Both cell types, in all animals, develop from the ectoderm, usually as sister cells or cousins derived from a common precursor. Thus, in vertebrates, the neurons and glial cells of the central nervous system (including the spinal cord, the brain, and the retina of the eye) derive from the part of the ectoderm that rolls up to form the neural tube, while those of the peripheral nervous system derive mainly from the neural crest (Figure 21-91).

Figure 21-91. Diagram of a 2-day chick embryo, showing the origins of the nervous system.

Figure 21-91

Diagram of a 2-day chick embryo, showing the origins of the nervous system. The neural tube (light green) has already closed, except at the tail end, and lies internally, beneath the ectoderm, of which it was originally a part (see Figure 21-76). The (more...)

The neural tube, with which we shall be mainly concerned, consists initially of a single-layered epithelium (Figure 21-92). The epithelial cells are the progenitors of the neurons and glia. As these cell types are generated, the epithelium becomes thickened and transformed into a more complex structure. Progenitor and, later, glial cells maintain the cohesiveness of the epithelium and form a scaffolding that spans its thickness. Along and between these tall cells, like animals amid the trees of the forest, the new-born neurons migrate, find their resting places, mature, and send out their axons and dendrites (Figure 21-93).

Figure 21-92. Formation of the neural tube.

Figure 21-92

Formation of the neural tube. The scanning electron micrograph shows a cross section through the trunk of a 2-day chick embryo. The neural tube is about to close and pinch off from the ectoderm; at this stage it consists (in the chick) of an epithelium (more...)

Figure 21-93. Migration of immature neurons.

Figure 21-93

Migration of immature neurons. Before sending out axons and dendrites, newborn neurons often migrate from their birthplace and settle in some other location. The diagrams are based on reconstructions from sections of the cerebral cortex of a monkey (part (more...)

Signal proteins secreted from the ventral and dorsal sides of the neural tube act as opposing morphogens, causing neurons born at different dorsoventral levels to express different gene regulatory proteins (Figure 21-94). There are differences along the head-to-tail axis as well, reflecting the anteroposterior pattern of expression of Hox genes and the actions of yet other morphogens. Moreover, neurons continue to be generated in each region of the central nervous system over many days, weeks, or even months, and this gives rise to still greater diversity, because the cells adopt different characters according to their “birthday”—the time of the terminal mitosis that marks the beginning of neuronal differentiation (Figure 21-95).

Figure 21-94. A schematic cross section of the spinal cord of a chick embryo, showing how cells at different levels along the dorsoventral axis express different gene regulatory proteins.

Figure 21-94

A schematic cross section of the spinal cord of a chick embryo, showing how cells at different levels along the dorsoventral axis express different gene regulatory proteins. (A) Signals that direct the dorsoventral pattern: Sonic hedgehog protein from (more...)

Figure 21-95. Programmed production of different types of neurons at different times from dividing progenitors in the cerebral cortex of the brain of a mammal.

Figure 21-95

Programmed production of different types of neurons at different times from dividing progenitors in the cerebral cortex of the brain of a mammal. Close to one face of the cortical neuroepithelium, progenitor cells divide repeatedly, in stem-cell fashion, (more...)

The Character Assigned to a Neuron at Its Birth Governs the Connections It Will Form

The differences of gene expression modulate the characters of the neurons and help to cause them to make connections with different partners. In the spinal cord, for example, ventrally located clusters of cells express genes of the Islet/Lim homeobox family (coding for gene regulatory proteins) and develop as motor neurons, sending out axons to connect with specific subsets of muscles—different muscles according to the particular Islet/Lim family members expressed. If the pattern of gene expression is artificially altered, the neurons project to different target muscles.

The different destinations reflect different pathway choices that the axons make as they grow out from the nerve cell body, as well as their selective recognition of different target cells at the end of the journey. In the dorsal part of the spinal cord lie neurons that receive and relay sensory information from sensory neurons in the periphery of the body. In intermediate positions, there are various other classes of interneurons, connecting specific sets of nerve cells to one another. Some send their axons dorsally, others ventrally; some up toward the head, others down toward the tail, still others across the floor of the neural tube to the other side of the body (Figure 21-96). In a timelapse film where the developing neurons are stained with a fluorescent dye, one can watch the movements of the growing tips of the axons as they extend: one is reminded of the lights of rush-hour traffic at night, as the cars streak along a network of highways, turning this way or that at busy junctions, each one making its own choice of route.

Figure 21-96. Growing axons in the developing spinal cord of a 3-day chick embryo.

Figure 21-96

Growing axons in the developing spinal cord of a 3-day chick embryo. The drawing shows a cross section stained by the Golgi technique. Most of the neurons, apparently, have as yet only one elongated process—the future axon. An irregularly shaped (more...)

How are these complex movements guided? Before attempting an answer, we must examine more closely the structure of the growing neuron.

Each Axon or Dendrite Extends by Means of a Growth Cone at Its Tip

A typical neuron sends out one long axon, projecting toward a distant target to which signals are to be delivered, and several shorter dendrites, on which it mainly receives incoming signals from axon terminals of other neurons. Each process extends by growth at its tip, where an irregular, spiky enlargement is seen. This structure, called the growth cone, appears to be crawling through the surrounding tissue, trailing a slender axon or dendrite behind it (see Figure 21-96). The growth cone comprises both the engine that produces the movement and the steering apparatus that directs the tip of each process along the proper path (see Figure 16-99).

Much of what we know about the properties of growth cones has come from studies in tissue or cell culture. One can watch as a neuron begins to put out its processes, all at first alike, until one of the growth cones puts on a sudden turn of speed, identifying its process as the axon, with its own axon-specific set of proteins (Figure 21-97). The contrast between axon and dendrite established at this stage involves polarized intracellular transport of different materials into the two types of process. As a result, they will grow out for different distances, follow different paths, and play different parts in synapse formation.

Figure 21-97. Formation of axon and dendrites in culture.

Figure 21-97

Formation of axon and dendrites in culture. A young neuron has been isolated from the brain of a mammal and put to develop in culture, where it sends out processes. One of these processes, the future axon, has begun to grow out faster than the rest (the (more...)

The growth cone at the end of a typical growing nerve cell process—either axon or dendrite—moves forward at a speed of about 1 mm per day, continually probing the regions that lie ahead and on either side by putting out filopodia and lamellipodia. When such a protrusion contacts an unfavorable surface, it withdraws; when it contacts a more favorable surface, it persists longer, steering the growth cone as a whole to move in that direction. In this way the growth cone can be guided by subtle variations in the surface properties of the substrata over which it moves. At the same time, it is sensitive to diffusible chemotactic factors in the surrounding medium, which can also encourage or hinder its advance. These behaviors depend on the cytoskeletal machinery inside the growth cone, as discussed in Chapter 16. A multitude of receptors in the growth cone membrane detect the external signals and, through the agency of intracellular regulators such as the monomeric GTPases Rho and Rac, control the assembly and disassembly of actin filaments and other components of the machinery of cell movement.

The Growth Cone Pilots the Developing Neurite Along a Precisely Defined Path in vivo

In living animals, growth cones generally travel toward their targets along predictable, stereotyped routes, exploiting a multitude of different cues to find their way, but always requiring a substratum of extracellular matrix or cell surface to crawl over. Often, growth cones take routes that have been pioneered by other neurites, which they follow by contact guidance. As a result, nerve fibers in a mature animal are usually found grouped together in tight parallel bundles (called fascicles or fiber tracts). Such crawling of growth cones along axons is thought to be mediated by homophilic cell-cell-adhesion molecules—membrane glycoproteins that help a cell displaying them to stick to any other cell that also displays them. As discussed in Chapter 19, two of the most important classes of such molecules are those that belong to the immunoglobulin superfamily, such as N-CAM, and those of the Ca2+-dependent cadherin family, such as N-cadherin. Members of both families are generally present on the surfaces of growth cones, of axons, and of various other cell types that growth cones crawl over, including glial cells in the central nervous system and muscle cells in the periphery of the body. The human genome contains more than 100 cadherin genes, for example, and most of them are expressed in the brain (see Figure 19-28). Different sets of cell-cell adhesion molecules, acting in varied combinations, provide a mechanism for selective neuronal guidance and recognition. Growth cones also migrate over components of the extracellular matrix. Some of the matrix molecules, such as laminin, favor axon outgrowth, while others, such as chondroitin sulfate proteoglycans, discourage it.

Growth cones are guided by a succession of different cues at different stages of their journey, and the stickiness of the substratum is not the only thing that matters. Another important part is played by chemotactic factors, secreted from cells that act as beacons at strategic points along the path—some attracting, others repelling. The trajectory of commissural axons—those that cross from one side of the body to the other—provides a beautiful example of how a combination of guidance signals can specify a complex path. Commissural axons are a general feature of bilaterally symmetrical animals, because the two sides of the body have to be neurally coordinated. Worms, flies and vertebrates use closely related mechanisms to guide their outgrowth.

In the developing spinal cord of a vertebrate, for example, a large number of neurons send their axonal growth cones ventrally toward the floor plate—a specialized band of cells forming the ventral midline of the neural tube (see Figure 21-96). The growth cones cross the floor plate and then turn abruptly through a right angle to follow a longitudinal path up toward the brain, parallel to the floor plate but never again crossing it (Figure 21-98A). The first stage of the journey depends on a concentration gradient of the protein netrin, secreted by the cells of the floor plate: the commissural growth cones sniff their way toward its source. Netrin was purified from chick embryos, by assaying extracts of neural tissue for an activity that would attract commissural growth cones in a culture dish. Its sequence revealed that it was the vertebrate homolog of a protein already known from C. elegans, through genetic screens for mutant worms with misguided axons—called unc mutants because they move in an uncoordinated fashion. One of the unc genes, unc-6, codes for the homolog of netrin. Another, unc-40, codes for its transmembrane receptor; and this too has a vertebrate homolog, called DCC that is expressed in the commissural neurons and mediates their response to the netrin gradient.

Figure 21-98. The guidance of commissural axons.

Figure 21-98

The guidance of commissural axons. (A) The pathway taken by commissural axons in the embryonic spinal cord of a vertebrate. (B) The signals that guide them. The growth cones are first attracted to the floorplate by netrin, which is secreted by the floor-plate (more...)

The receptors on each growth cone determine the route it will take: non-commissural neurons in the neural tube, lacking DCC, are not attracted to the floorplate; and neurons expressing a different netrin receptor—called Unc-5H in vertebrates (with a counterpart Unc-5 in the worm)—are actively repelled by the floorplate and send their axons instead toward the roof plate.

Growth Cones Can Change Their Sensibilities as They Travel

If commissural growth cones are attracted to the floor plate, why do they cross it and emerge on the other side, instead of staying in the attractive territory? And having crossed it, why do they never veer back onto it again? The likely answer lies in another set of molecules, several of which are also conserved between vertebrates and invertebrates. Studies of Drosophila mutants with misguided commissural axons first identified three of the key proteins: Slit, Roundabout, and Commissureless.

Slit, like netrin, is produced by midline cells of the developing fly, while its receptor, Roundabout, is expressed in the commissural neurons. Slit, acting on Roundabout, has an effect exactly opposite to that of netrin: it repels the growth cones, blocking entry to the midline territory. Commissureless, however, apparently inhibits Roundabout expression initially so as to make the growth cones blind to this “keep-out” signal. Commissural growth cones supplied with Commissureless protein advance to the midline; as they cross it, they seem to lose their blindfold of Commissureless protein and begin to be repelled. Emerging on the far side, they now have functional Roundabout on their surfaces and are thereby prohibited from re-entry.

In vertebrates, a similar mechanism may operate. Commissural growth cones are at first attracted to the midline, and then change their surface receptor proteins as they cross; in this way they may switch their sensibilities, gaining sensitivity to repulsion by Slit—which is expressed in the floor plate—and losing sensitivity to attraction by netrin. Repulsion from the midline now prevents them from straying back across it. At the same time, the growth cones apparently become sensitive to another set of repulsive signals, in the form of proteins called semaphorins, which prevent them from traveling back up into the dorsal regions of the spinal cord. Trapped between the two sets of repulsive signals, the growth cones have no choice but to travel in a narrow track, running parallel to the floor plate but never re-entering it (Figure 21-98B).

Target Tissues Release Neurotrophic Factors That Control Nerve Cell Growth and Survival

Eventually, axonal growth cones reach the target region where they must halt and make synapses. The neurons that sent out the axons can now begin to communicate with their target cells. Although synapses generally transmit signals in one direction, from axon to either dendrite or muscle, the developmental communications are a two-way affair. Signals from the target tissue not only regulate which growth cones shall synapse where (as we discuss below), but also how many of the innervating neurons shall survive.

Most types of neurons in the vertebrate central and peripheral nervous system are produced in excess; up to 50% or more of them then die soon after they reach their target, even though they appear perfectly normal and healthy up to the time of their death. About half of all the motor neurons that send axons to skeletal muscle, for example, die within a few days after making contact with their target muscle cells. A similar proportion of the sensory neurons that innervate the skin die after their growth cones have arrived there.

This large-scale death of neurons is thought to reflect the outcome of a competition. Each type of target cell releases a limited amount of a specific neurotrophic factor that the neurons innervating that target require to survive. The neurons apparently compete for the factor, and those that do not get enough die by programmed cell death. If the amount of target tissue is increased—for example, by grafting an extra limb bud onto the side of the embryo—more limb-innervating neurons survive; conversely, if the limb bud is cut off, the limb-innervating neurons all die. In this way, although individuals may vary in their bodily proportions, they always retain the right number of motor neurons to innervate all their muscles and the right number of sensory neurons to innervate their whole body surface. The seemingly wasteful strategy of overproduction followed by death of surplus cells operates in almost every region of the nervous system. It provides a simple and effective means to adjust each population of innervating neurons according to the amount of tissue requiring innervation.

The first neurotrophic factor to be identified, and still the best characterized, is known simply as nerve growth factor, or NGF—the founding member of the neurotrophin family of signal proteins. It promotes the survival of specific classes of sensory neurons derived from the neural crest and of sympathetic neurons (a subclass of peripheral neurons that control contractions of smooth muscle and secretion from exocrine glands), NGF is produced by the tissues that these neurons innervate. When extra NGF is provided, extra sensory and sympathetic neurons survive, just as if extra target tissue were present. Conversely, in a mouse with a mutation that knocks out the NGF gene or its receptor (a transmembrane tyrosine kinase called TrkA), almost all sympathetic neurons and the NGF-dependent sensory neurons are lost. There are many neurotrophic factors, only a few of which belong to the neurotrophin family, and they act in different combinations to promote survival of different classes of neurons.

NGF and its relatives have an additional role: besides acting on the nerve cell as a whole to control its survival, they regulate the outgrowth of axons and dendrites (Figure 21-99). These can even act locally on just one part of the tree of nerve cell processes, promoting or pruning the growth of individual branches: a growth cone exposed to NGF shows an immediate increase of motility. Conversely, an axon branch that is deprived of NGF, while the rest of the neuron continues to be bathed in the factor, dies back.

Figure 21-99. NGF effects on neurite outgrowth.

Figure 21-99

NGF effects on neurite outgrowth. Dark-field photomicrographs of a sympathetic ganglion cultured for 48 hours with (above) or without (below) NGF. Neurites grow out from the sympathetic neurons only if NGF is present in the medium. Each culture also contains (more...)

The peripheral action of NGF continues to be important after the phase of neuronal death. In the skin, for example, it controls the branching of sensory nerve fibers, ensuring not only that the whole body surface becomes innervated during development but also that it recovers its innervation after damage.

Neuronal Specificity Guides the Formation of Orderly Neural Maps

In many cases, axons originating from neurons of a similar type but located in different positions come together for the journey and arrive at the target in a tight bundle. There they disperse again, to terminate at different sites in the target territory.

The projection from the eye to the brain provides an important example. The neurons in the retina that convey visual information back to the brain are called retinal ganglion cells. There are more than a million of them, each one reporting on a different part of the visual field. Their axons converge on the optic nerve head at the back of the eye and travel together along the optic stalk into the brain. Their main site of termination, in most vertebrates other than mammals, is the optic tectum—a broad expanse of cells in the midbrain. In connecting with the tectal neurons, the retinal axons distribute themselves in a predictable pattern according to the arrangement of their cell bodies in the retina: ganglion cells that are neighbors in the retina connect with target cells that are neighbors in the tectum. The orderly projection creates a map of visual space on the tectum (Figure 21-100).

Figure 21-100. The neural map from eye to brain in a young zebrafish.

Figure 21-100

The neural map from eye to brain in a young zebrafish. (A) Diagrammatic view, looking down on the top of the head. (B) Fluorescence micrograph. Fluorescent tracer dyes have been injected into each eye—red into the anterior part, green into the (more...)

Orderly maps of this sort are found in many brain regions. In the auditory system, for example, neurons project from the ear to the brain in a tonotopic order, creating a map in which brain cells receiving information about sounds of different pitch are ordered along a line, like the keys of a piano. And in the somatosensory system, neurons conveying information about touch map onto the cerebral cortex so as to mark out a “homunculus”—a small, distorted, two dimensional image of the body surface (Figure 21-101).

Figure 21-101. A map of the body surface in the human brain.

Figure 21-101

A map of the body surface in the human brain. The surface of the body is mapped onto the somatosensory region of the cerebral cortex by an orderly system of nerve cell connections, such that sensory information from neighboring body sites is delivered (more...)

The retinotopic map of visual space in the optic tectum is the best characterized of all these maps. How does it arise? In principle, the growth cones could be physically channeled to different destinations as a consequence of their different starting positions, like drivers on a multilane highway where it is forbidden to change lanes. This possibility was tested in the visual system by a famous experiment in the 1940s. If the optic nerve of a frog is cut, it will regenerate. The retinal axons grow back to the optic tectum, restoring normal vision. If, in addition, the eye is rotated in its socket at the time of cutting of the nerve, so as to put originally ventral retinal cells in the position of dorsal retinal cells, vision is still restored, but with an awkward flaw: the animal behaves as though it sees the world upside down and left-right inverted. This is because the misplaced retinal cells make the connections appropriate to their original, not their actual, positions. It seems that the cells have positional values—position-specific biochemical properties representing records of their original location. As a result, cells on opposite sides of the retina are intrinsically different, just as the motor neurons in the spinal cord that project to different muscles are intrinsically different.

Such nonequivalence among neurons is referred to as neuronal specificity. It is this intrinsic characteristic that guides the retinal axons to their appropriate target sites in the tectum. Those target sites themselves are distinguishable by the retinal axons because the tectal cells also carry positional labels. Thus the neuronal map depends on a correspondence between two systems of positional markers, one in the retina and the other in the tectum.

Axons From Different Regions of the Retina Respond Differently to a Gradient of Repulsive Molecules in the Tectum

Axons from the nasal retina (the side closest to the nose) project to the posterior tectum, and axons from the temporal retina (the side farthest from the nose) project to the anterior tectum, with intermediate regions of retina projecting to intermediate regions of tectum. When nasal and temporal axons are allowed to grow out over a carpet of anterior or posterior tectal membranes in a culture dish, they also show selectivity (Figure 21-102). Temporal axons strongly prefer the anterior tectal membranes, as in vivo, whereas nasal axons either prefer posterior tectal membranes, or show no preference (depending on the species of animal). The key difference between anterior and posterior tectum appears to be a repulsive factor on the posterior tectum, to which temporal retinal axons are sensitive but nasal retinal axons are not: if a temporal retinal growth cone touches posterior tectal membrane, it collapses its filopodia and withdraws.

Figure 21-102. Selectivity of retinal axons growing over tectal membranes.

Figure 21-102

Selectivity of retinal axons growing over tectal membranes. (A) A photograph of the experimental observation. (B) A diagram of what is happening. The culture substratum has been coated with alternating stripes of membrane prepared either from posterior (more...)

Assays based on these phenomena in vitro have identified some of the molecules responsible. The repulsive factor on posterior tectal membrane seems to be partly or entirely comprised of ephrin A proteins, a subset of the family of GPI-linked proteins that act as ligands for the Eph family of tyrosine kinase receptors. In the mouse, two different ephrins are expressed to form an anterior-to-posterior gradient on the tectal cells. Anterior cells have little or no ephrin, cells in the center of the tectum express ephrin A2, and cells at the posterior edge of the tectum express ephrin A2 and ephrin A5. Thus there is a gradient of ephrin expression across the tectum. Meanwhile, the incoming axons express Eph receptors, also in a gradient: temporal axons express high Eph levels, making them sensitive to repulsion by ephrin A, whereas nasal axons express low Eph levels.

This system of signals and receptors is enough to produce an orderly map, if we make one further assumption—an assumption supported by experiments in vivo: that the retinal axons somehow interact with one another and compete for tectal territory. Thus, temporal axons are restricted to anterior tectum, and drive nasal axons off it; nasal axons, consequently, are restricted to posterior tectum. Between the extremes, a balance is struck, creating a smooth map of the temporo-nasal axis of the retina onto the anteroposterior axis of the tectum. Ephrin gene knockout studies in the mouse are consistent with this picture, although they also indicate that additional cues help to guide the pattern of retino-tectal projections. Moreover, the map also has to be patterned along the dorsoventral axis; it is thought that this depends on similar mechanisms, perhaps even involving some of the same molecules.

Diffuse Patterns of Synaptic Connections Are Sharpened by Activity-Dependent Remodeling

In a normal animal the retinotectal map is initially fuzzy and imprecise: the system of matching markers we have just described is enough to define the broad layout of the map, but not sufficient to specify its fine details. Studies in frogs and fish show that each retinal axon at first branches widely in the tectum and makes a profusion of synapses, distributed over a large area of tectum that overlaps with the territories innervated by other axons. These territories are subsequently trimmed back by selective elimination of synapses and retraction of axon branches. This is accompanied by the formation of new sprouts, through which each axon develops a denser distribution of synapses in the territory that it retains.

A central part in this remodeling and refinement of the map is played by two competition rules that jointly help to create spatial order: (1) axons from separate regions of retina, which tend to be excited at different times, compete to dominate the available tectal territory, but (2) axons from neighboring sites in the retina, which tend to be excited at the same time, innervate neighboring territories in the tectum because they collaborate to retain and strengthen their synapses on shared tectal cells (Figure 21-103). The mechanism underlying both these rules depends on electrical activity and signaling at the synapses that are formed. If all action potentials are blocked by a toxin that binds to voltage-gated Na+ channels, synapse remodeling is inhibited and the map remains fuzzy.

Figure 21-103. Sharpening of the retinotectal map by synapse elimination.

Figure 21-103

Sharpening of the retinotectal map by synapse elimination. At first the map is fuzzy because each retinal axon branches widely to innervate a broad region of tectum overlapping the regions innervated by other retinal axons. The map is then refined by (more...)

The phenomenon of activity-dependent synapse elimination is encountered in almost every part of the developing vertebrate nervous system. Synapses are first formed in abundance and distributed over a broad target field; then the system of connections is pruned back and remodeled by competitive processes that depend on electrical activity and synaptic signaling. The elimination of synapses in this way is distinct from the elimination of surplus neurons by cell death, and it occurs after the period of normal neuronal death is over.

Much of what we know about the cellular mechanisms of synapse formation and elimination comes from experiments on the innervation of skeletal muscle in vertebrate embryos. A two-way exchange of signals between the nerve axon terminals and the muscle cells controls the initial formation of synapses. At sites of contact, acetylcholine receptors are clustered in the muscle cell membrane and the apparatus for secretion of this neurotransmitter becomes organized in the axon terminals (see Chapter 11). Each muscle cell at first receives synapses from several neurons; but in the end, through a process that typically takes a couple of weeks, it is left innervated by only one. The synapse retraction again depends on synaptic communication: if synaptic transmission is blocked by a toxin that binds to the acetylcholine receptors in the muscle cell membrane, the muscle cell retains its multiple innervation beyond the normal time of elimination.

Experiments on the musculoskeletal system, as well as in the retinotectal system, suggest that it is not only the amount of electrical activity at a synapse that is important for its maintenance, but also its temporal coordination. Whether a synapse is strengthened or weakened seems to depend critically on whether or not activity in the presynaptic cell is synchronized with activity of the other presynaptic cells synapsing on the same target (and thus also synchronized with activity of the target cell itself).

These and many other findings have suggested a simple interpretation of the competition rules for synapse elimination in the retinotectal system (Figure 21-104). Axons from different parts of the retina fire at different times and so compete. Each time one of them fires, the synapse(s) made by the other on a shared tectal target cell are weakened, until one of the axons is left in sole command of that cell. Axons from neighboring retinal cells, on the other hand, tend to fire in synchrony with one another: they therefore do not compete but instead maintain synapses on shared tectal cells, creating a precisely ordered map in which neighboring cells of the retina project to neighboring sites in the tectum.

Figure 21-104. Synapse modification and its dependence on electrical activity.

Figure 21-104

Synapse modification and its dependence on electrical activity. Experiments in several systems indicate that synapses are strengthened or weakened by electrical activity according to the rule shown in the diagram. The underlying principle appears to be (more...)

Experience Molds the Pattern of Synaptic Connections in the Brain

The phenomenon that we have just described is summed up in the catch-phrase that “neurons that fire together, wire together”. The same firing rule relating synapse maintenance to neural activity helps to organize our developing brains in the light of experience.

In the brain of a mammal, axons relaying inputs from the two eyes are brought together in a specific cell layer in the visual region of the cerebral cortex. Here, they form two overlapping maps of the external visual field, one as perceived through the right eye, the other as perceived through the left. Although there is some evidence of a tendency for right- and left-eye inputs to be segregated even before synaptic communication begins, a large proportion of the axons carrying information from the two eyes at early stages synapse together on shared cortical target cells. A period of early signaling activity, however, occurring spontaneously and independently in each retina even before vision begins, leads to a clean segregation of inputs, creating stripes of cells in the cortex that are driven by inputs from the right eye alternating with stripes that are driven by inputs from the left eye (Figure 21-105). The firing rule suggests a simple interpretation: a pair of axons bringing information from neighboring sites in the left eye will frequently fire together, and therefore wire together, as will a pair of axons from neighboring sites in the right eye; but a right-eye axon and a left-eye axon will rarely fire together, and will instead compete. Indeed, if activity from both eyes is silenced using drugs that block action potentials or synaptic transmission, the inputs fail to segregate correctly.

Figure 21-105. Ocular dominance columns in the visual cortex of a monkey's brain, and their sensitivity to visual experience.

Figure 21-105

Ocular dominance columns in the visual cortex of a monkey's brain, and their sensitivity to visual experience. (A) Normally, stripes of cortical cells driven by the right eye alternate with stripes, of equal width, driven by the left eye. The stripes (more...)

Maintenance of the pattern of connections is extraordinarily sensitive to experience early in life. If, during a certain critical period (ending at about the age of five years in humans), one eye is kept covered for a time so as to deprive it of visual stimulation, while the other eye is allowed normal stimulation, the deprived eye loses its synaptic connections to the cortex and becomes almost entirely, and irreversibly, blind. In accordance with what the firing rule would predict, a competition has occurred in which synapses in the visual cortex made by inactive axons are eliminated while synapses made by active axons are consolidated. In this way cortical territory is allocated to axons that carry information and is not wasted on those that are silent.

In establishing the nerve connections that enable us to see, it is not only the quantity of visual stimulation that is important, but also its temporal coordination. For example, the ability to see depth—stereo vision—depends on cells in other layers of the visual cortex that receive inputs relayed from both eyes at once, conveying information about the same part of the visual field as seen from two slightly different angles. These binocularly driven cells allow us to compare the view through the right eye with that through the left so as to derive information about the relative distances of objects from us. If, however, the two eyes are prevented during the critical period from ever seeing the same scene at the same time—for example, by covering first one eye and then the other on alternate days or simply as a consequence of a childhood squint—almost no binocularly driven cells are retained in the cortex, and the capacity for stereo perception is irretrievably lost. Evidently, in accordance with the firing rule, the inputs from each eye to a binocularly driven neuron are maintained only if the two inputs are frequently triggered to fire in synchrony, as occurs when the two eyes look together at the same scene.

Adult Memory and Developmental Synapse Remodeling May Depend on Similar Mechanisms

We saw in Chapter 11 that synaptic changes underlying memory in at least some parts of the adult brain, notably the hippocampus, hinge on the behavior of a particular type of receptor for the neurotransmitter glutamate—the NMDA receptor. Ca2+ flooding into the postsynaptic cell through the channels opened by this receptor triggers lasting changes in the strengths of the synapses on that cell, affecting the presynaptic as well as the postsynaptic structures. The changes that are induced by the NMDA-dependent mechanism in the adult brain obey rules closely akin to the developmental firing rule: events in the external world that cause two neurons to be active at the same time, or in quick succession, favor the making or strengthening of synapses between them. This condition, called the Hebb rule, has been suggested to be the fundamental principle underlying associative learning.

Is it possible, then, that adult learning and the more drastic forms of synaptic plasticity seen during development both reflect the operation of the same basic machinery of synapse adjustment? There are many hints that it may be so. For example, inhibitors that specifically block activation of the NMDA receptor have been found to interfere with the refinement and remodeling of synaptic connections in the developing visual system. But the question is still open. The molecular basis of the processes of synapse remodeling through which experience molds our brains remains one of the central challenges that the nervous system presents to cell biology.


The development of the nervous system proceeds in three phases: first, nerve cells are generated through cell division; then, having ceased dividing, they send out axons and dendrites to form profuse synapses with other, remote cells so that communication can begin; last, the system of synaptic connections is refined and remodeled according to the pattern of electrical activity in the neural network.

The neurons, and the glial cells that always accompany them, are generated from ectodermal precursors, and those born at different times and places express different sets of genes, which help to determine the connections they will form. Axons and dendrites grow out from the neurons by means of growth cones, which follow specific pathways delineated by signals along the way. Structures such as the floor plate of the embryonic spinal cord secrete both chemoattractants and chemorepellents, to which growth cones from different classes of neurons respond differently. On reaching their target area, the axons terminate selectively on a subset of the accessible cells, and in many parts of the nervous system neural maps are set up—orderly projections of one array of neurons onto another. In the retinotectal system, the map is based on the matching of complementary systems of position-specific cell-surface markers—ephrins and Eph receptors—possessed by the two sets of cells.

After the growth cones have reached their targets and initial connections have formed, two major sorts of adjustment occur. First, many of the innervating neurons die as a result of a competition for survival factors such as NGF (nerve growth factor) secreted by the target tissue. This cell death adjusts the quantity of innervation according to the size of the target. Second, individual synapses are pruned away in some places, reinforced in others, so as to create a more precisely ordered pattern of connections. This latter process depends on electrical activity: synapses that are frequently active are reinforced, and different neurons contacting the same target cell tend to maintain their synapses on the shared target only if they are both frequently active at the same time. In this way the structure of the brain can be adjusted to reflect the connections between events in the external world. The underlying molecular mechanism of this synaptic plasticity may be similar to that responsible for the formation of memories in adult life.

Image ch21f76
Image ch16f99
Image ch19f28

By agreement with the publisher, this book is accessible by the search feature, but cannot be browsed.

Copyright © 2002, Bruce Alberts, Alexander Johnson, Julian Lewis, Martin Raff, Keith Roberts, and Peter Walter; Copyright © 1983, 1989, 1994, Bruce Alberts, Dennis Bray, Julian Lewis, Martin Raff, Keith Roberts, and James D. Watson .
Bookshelf ID: NBK26814