Molecular Biology of the Cell - Vol. 3 - Alberts B., Bray D., Lewis J., Raff M., Roberts K., Watson J. 1994

From Cells to Multicellular Organisms
The Nervous System
Formation and Elimination of Synapses

The meeting of a growth cone with a target Cell is a pivotal moment in neuronal development: both the growth cone and the target cell undergo a transformation that establishes a synaptic connection. However, development does not stop there—many of the initially formed synapses are subsequently eliminated, while new ones are established elsewhere on the same target cell. Such local remodeling of synaptic connections provides a mechanism for correcting wiring errors and executing "fine-tuning": the initial pattern of connections is roughly mapped out by guidance cues that steer the migration of growth cones along specific pathways toward their target Cells; preliminary synaptic junctions then form, allowing pre- and postsynaptic cells to interact; and finally, these initial connections are reviewed and refined through mechanisms utilizing both neurotropic factors and electrical signals in the form of action potentials and synaptic potentials. Consequently, external stimuli capable of eliciting electrical activity in the Nervous system can influence The Development of neural circuits.

In this section, we examine synapse formation at THE MOLECULAR LEVEL, the rules governing whether a synapse is formed or eliminated, and The Role of electrical activity in regulating these processes. We begin with the synapses between motor Neurons and Skeletal Muscle cells, as they are the best understood.

19.8.1. Synaptic contact induces specialization in these Regions of the growing axon and target cell for the function of signal transmission [62]

The Cytology/cytology/16.html">Early stages of neuromuscular synapse formation are most readily observed in culture. Here, one can observe that a significant portion of the molecular machinery for synaptic transmission exists even before the growth cone reaches the muscle cell. As the growth cone advances, electrical excitation of the neuronal cell body causes it to release small amounts of acetylcholine (Fig. 19-75). The growth cone membrane already contains voltage-gated calcium channels that couple electrical excitation to secretion; these channels also serve to propagate nerve impulses along the embryonic neurite (which initially lacks sodium channels). Even before the muscle cell is innervated, it already possesses acetylcholine receptors (of the embryonic type) and can respond to acetylcholine by depolarizing and contracting.

Relatively inefficient synaptic transmission can be observed just minutes after the initial contact between the growth cone and the muscle cell. However, for a mature synapse to form, both the growth cone and the target cell must undergo Structural and Biochemical specialization—a process that typically takes several days. The growth cone ceases its movement, synaptic vesicles accumulate within it, and specialized "active zones" form at a specific site for the rapid and highly localized release of acetylcholine (Section 19.3.3). Acetylcholine receptors on the muscle cell surface become concentrated at the synaptic site, whereas their density diminishes in other areas of The Plasma Membrane. How is this redistribution of neurotransmitter receptors achieved? This question applies not only to muscle cells but also to neurons: for successful signal transmission and information Processing, neurons must likewise concentrate specific types of receptors and Ion Channels in distinct regions of the plasma membrane.

Class="center">

Fig. 19-75. Schematic diagram of an experiment demonstrating that portions of acetylcholine are released from the motor neuron growth cone in response to stimulation of The Cell body. Minute quantities of the released neurotransmitter are detected by measuring its effect on the current flowing through a patch of muscle fiber membrane that covers the tip of a micropipette and contains numerous acetylcholine receptors. Acetylcholine is released from the growth cone in much smaller quantities and with less regularity than from a mature synaptic terminal.

19.8.2. Acetylcholine receptors diffuse within the muscle Cell Membrane and cluster at the site of synapse formation [63]

In a mature muscle cell, the concentration of acetylcholine receptors at the synapse is over a thousandfold higher than in other membrane regions. Fluorescence photobleaching recovery experiments (Section 6.2.9) show that the receptors at the synapse are "anchored" and cannot move freely within the plane of the membrane. By contrast, in an uninnervated embryonic muscle cell, receptors are distributed across the entire surface and are capable of more free diffusion. When a motor neuron axon forms a contact with a muscle cell, acetylcholine receptors begin to aggregate in the membrane domain underlying the axon terminal; newly synthesized receptors are now also inserted into the membrane predominantly in the region of the developing synapse (Fig. 19-76). The receptors become fixed in place—possibly As a result of mutual adhesion, or perhaps due to binding with the Cytoskeleton or the Extracellular matrix. Some important insights into how an axon "selects" the future synaptic site have come from studies of Neuromuscular Junction regeneration.

19.8.3. The site of neuromuscular contact is distinguished by a stable specialization of the basal lamina [64]

Every muscle cell is surrounded by a basal lamina (see Figs. 19-16 and 19-18, A). In the event of severe injury, a muscle fiber degenerates and dies, and its cellular debris is cleared by macrophages. However, the basal lamina remains intact and acts as a "scaffold" from which a new muscle fiber can regenerate from surviving stem cells (Section 19.7.3 / Section 17.6.3). Even when both the muscle fiber and the nerve terminal are destroyed, the site of the former neuromuscular contact can still be identified by the uneven surface of the basal lamina in that area. This synaptic basal lamina possesses unique chemical properties, and Antibodies can be generated that selectively bind to its surface. Interestingly, it is precisely the synaptic basal lamina that determines the localization of the remaining synaptic components. Structure/19.html">The Importance of the basal lamina in neuromuscular junction formation was demonstrated in a series of experiments on amphibians. Following the simultaneous destruction of the nerve and muscle cell, leaving only an empty sleeve of basal lamina, it is easy to verify that the synaptic region of the basal lamina specifically retains molecules of acetylcholinesterase, which hydrolyzes the acetylcholine released by nerve terminals in a normal synapse. Furthermore, the basal lamina also anchors the axon terminal at the synaptic site; if only the muscle cell is destroyed, the terminal remains attached to the basal lamina for many days. Conversely, removing the basal lamina with collagenase causes the axon terminal to detach even when the muscle cell is preserved.

Fig. 19-76. Clustering of acetylcholine receptors in the region of the developing muscle fiber membrane where a synapse forms between the motor axon terminal and the cell. This clustering is partly driven by the lateral diffusion of receptors from neighboring areas of the muscle fiber membrane and partly by the insertion of newly synthesized receptors directly into the membrane at this site. These processes appear to be independent of neurotransmitter release from the nerve terminal, as they occur even in the presence of agents that block action potentials in the nerve cell, and even when high concentrations of $\alpha$-bungarotoxin are present in the extracellular medium (a snake venom component that binds to acetylcholine receptors and blocks their interaction with acetylcholine). Receptors accumulated at the synapse are effectively trapped—their turnover rate here is much lower than in other membrane domains, with a half-life / lifetime reaching five days or more.

Indeed, it appears that the basal lamina itself is capable of directing the regeneration of the axon terminal. This was demonstrated in the following experiment: the nerve and muscle cell are destroyed, after which the nerve is allowed to regenerate. Although the basal lamina sheath remains empty, the regenerating axon tracks down the original synaptic site and establishes a synaptic terminal there. In addition, the basal lamina controls the localization of acetylcholine receptors at the synaptic junction. If the nerve and muscle fiber are destroyed and the muscle is allowed to regenerate while nerve regeneration is blocked, the acetylcholine receptors synthesized by the regenerated muscle localize preferentially to the region of the former junction, despite the absence of the nerve (Fig. 19-77). As might be expected, extracts prepared from the basal lamina of the neuromuscular junction contain a protein called agrin, which promotes receptor aggregation in muscle cell cultures.

Evidently, when an axon contacts a muscle cell, it deposits (or induces the muscle cell to deposit) macromolecules such as agrin that stabilize the synaptic connection. However, the basal lamina alone is not sufficient for neuromuscular synapse formation: by no means all contacts between an axon and a muscle cell result in synapse formation, and not all newly formed synapses are entirely stable.

Fig. 19-77. Experiment demonstrating that the specific Properties of the neuromuscular junction basal lamina regulate the localization of other synaptic components.

19.8.4. The susceptibility of a muscle cell to synapse formation is regulated by its electrical activity [63, 65]

If a rat's nerve is severed and repositioned so that its cut end lies over an adjacent normal muscle, the severed axons will regenerate and grow across The surface of this muscle; however, as long as the normal innervation of this muscle remains intact, foreign axons will not make contact with individual cells of this muscle or form synapses on them. By contrast, if the nerve normally innervating the muscle is cut, striking changes ensue. Over the course of a few days, the METABOLISM and membrane properties of the muscle cells change—specifically, large numbers of new acetylcholine receptors are synthesized and inserted into the membrane, rendering the cell hypersensitive to acetylcholine. At the same time, the muscle cells become receptive to The formation of new synapses with "foreign" axons growing across the muscle surface. Although these axons preferentially target former synaptic sites, they can also form synapses on novel regions of the muscle cells. Once synapses are established, the uniform distribution of acetylcholine receptors disappears (just as during embryonic development)—they persist in high concentrations solely at the sites of synapse formation (Fig. 19-78).

Denervation of a muscle abolishes its normal neural stimulation: the changes described above are driven primarily by the lack of electrical activity in the muscle cells; this same lack of activity triggers the release of the axon-branching-stimulating factor mentioned earlier (Section 19.8.4). All these denervation effects, which render the muscle more receptive to synaptogenesis, can be reproduced by local anesthesia of an intact nerve, which blocks impulse conduction to the muscle. Conversely, if a denervated muscle is artificially stimulated via implanted electrodes, the sensitivity of extrajunctional membrane regions to acetylcholine is suppressed, as is the formation of new synapses. Normally, the electrical activity elicited by a neuron that has already formed a synapse "shields" the cell against unwanted additional innervation.

In much the same way, electrical activity regulates synapse elimination during development. In vertebrate embryos, an uninnervated muscle is contacted by multiple nerve terminals almost simultaneously, leading to the initial formation of numerous redundant synapses. The mature pattern of innervation, in which each muscle cell receives only a single synapse, is established through two distinct processes separated in time. The first is the death of excess motor neurons, and the second is the elimination of surplus axon branches (synapse elimination).

Fig. 19-78. Experiment on the rat soleus muscle demonstrating that a denervated muscle cell retains the capacity for synapse formation following the transplantation of a foreign nerve fiber. Note that denervation alters the distribution of acetylcholine receptors in the muscle cell membrane: newly synthesized receptors become distributed across the entire cell surface, although receptor concentration remains particularly high at the site of the former neuromuscular junction. Denervation also alters membrane electrical excitability due to the appearance of a novel class of voltage-gated channels in the membrane that are relatively resistant to tetrodotoxin.

19.8.5. Muscle electrical activity influences the survival of embryonic motor neurons [59, 66]

As noted previously, approximately 50% of embryonic motor neurons die shortly after establishing synaptic contacts with muscle cells. This loss of surplus neurons can be prevented by blocking neuromuscular transmission (e.g., with $\alpha$-bungarotoxin) or, conversely, exacerbated by subjecting the muscle to direct electrical stimulation. This suggests that muscle electrical activity regulates the production within the muscle of a neurotrophic factor essential for embryonic motor neuron survival. This factor may be identical to the putative factor thought to stimulate axonal growth toward a denervated muscle. When a muscle is rendered inactive by blockade of synaptic transmission or the absence of innervating axons, this factor is produced in large quantities as a signal that the cell requires innervation. Electrical activation of the muscle via artificial stimuli or spontaneous excitation of its innervating motor neurons suppresses factor production, resulting in the death of a fraction of immature embryonic motor neurons as they compete for the remaining scarce quantities.

19.8.6. Electrical Activity Regulates the Competitive Elimination of Synapses According to a “Rule of Activation” [61, 67]

Even after the death of half of the embryonic motor neurons, a large excess of synapses remains on the developing muscle. Each motor neuron branches extensively, forming synapses on several muscle cells, and each muscle cell is most commonly innervated by the processes of multiple neurons. To establish the pattern of connections characteristic of the adult Organism, all but one of the synapses on each muscle cell must be eliminated. The process of developmental synapse elimination has been well studied in the rat soleus muscle ($m.\text{ soleus}$), where at birth an individual muscle cell is innervated by an average of three motor neurons. Over the subsequent two to three weeks, each neuron retracts a significant portion of its terminal branches until each cell is innervated by a single, exclusive branch of one motor neuron (Fig. 19-79). If excess axon branches were eliminated at random, some muscle fibers might be completely deprived of innervation, while others would retain multiple synapses. The fact that only a single synapse remains on each cell implies that the process of synapse elimination is competitive. Indeed, the competitive elimination of redundant synapses throughout the nervous system represents one of the most critical mechanisms regulating the development of neural connections and, as we will see later, their subsequent modification in response to external stimuli. Although the molecular mechanisms underlying competitive synapse elimination remain elusive, competition in most cases appears to follow a few simple, general rules applicable to both neuromuscular and interneuronal synapses.

Fig. 19-79. Elimination of redundant synapses in mammalian skeletal muscle after birth. For clarity, only a fraction of the motor axon terminal branches are shown; in reality, a single axon in a mature muscle innervates hundreds of muscle fibers with its arborizations. All axonal branches innervating a single immature muscle fiber typically form synapses within the same small region of the muscle cell and compete until only one synapse remains.

Fig. 19-80. One of several molecular mechanisms that could underlie the “rule of activation.” According to the hypothesis presented here, the maintenance of a synapse depends on a nerve growth factor secreted by the postsynaptic cell. The secretion of this factor is triggered by local electrical stimulation when an Action Potential arrives at the synapse; at other times, the neurotransmitter is released spontaneously, but at a lower rate. The growth factor is taken up by the recently activated axon terminal via endocytosis along with membrane material immediately following transmitter release (Section 19.3.4). Terminals that take up the growth factor enlarge and accumulate substances that strengthen the synaptic connection. The more frequently the muscle is stimulated, the more rapidly the intracellular stores of the factor are depleted, and the less frequently spontaneous release occurs without stimulation. Consequently, inactive nerve endings competing with active ones fail to receive the required amount of growth factor, progressively shrink, and ultimately die. Two endings active at different times compete for a limited amount of growth factor contained within the postsynaptic cell. Because larger endings take up more growth factor and consequently become even larger, the final outcome of such competition may depend on a slight initial difference in terminal size. According to an alternative hypothesis, stimulation of the postsynaptic cell leads to the local release or Activation of a protease that degrades inactive synapses, whereas active synapses are somehow protected from the action of this protease.

First, although competition initially involves an element of chance, the final outcome is absolute: the neuron either survives or is completely eliminated. Second, competition typically occurs only between synapses located relatively close to one another on the same target cell. Thus, during normal development of a typical mammalian skeletal muscle fiber, incoming nerve endings form synapses within the same small region and compete among themselves until only one synapse remains. If the formation of multiple synapses on the same fiber is artificially induced at a distance of $1\text{ mm}$ or more apart, however, multiple innervation persists.

Third, and most importantly, competitive synapse elimination depends on electrical activity in both the axons and the cells they innervate. For example, it slows down if the excitation of the developing muscle is blocked by applying a local anesthetic to the nerve or $\alpha$-bungarotoxin to the neuromuscular junctions. In most systems studied to date, inactivation of a fraction of the innervating axons results in the remaining active axons taking control of a larger number of target cells. This suggests that in the vicinity of an active synapse, the stimulated target cell either secretes a factor that promotes the destruction of neighboring synapses or, conversely, ceases to synthesize a substance necessary for synapse maintenance.

Yet this presents a puzzling question: if synaptic stimulation by the target cell leads to synapse destruction, why do the very synapses through which the cell receives stimulation persist and strengthen? The answer apparently lies in the following rule: any excitation of the target cell strengthens those synapses where the presynaptic axon terminal was previously active, and promotes the elimination of those synapses where the terminal remained quiescent.

Fig. 19-81. Changes in the arrangement of synapses on rat submandibular ganglion neurons shortly after birth. Initially, each cell is innervated by multiple axons. These axons compete until synapse elimination leaves only a single axon. This sole axon then forms numerous synapses on the cell that no longer compete with one another. (After D. Purves, J.W. Lichtman, Physiol. Rev., 58, 821-862, 1978.)

Therefore, temporal relationships between axon and target cell activation play a decisive role; when several independently activated neurons form contacts with the same cell, each of these neurons strives to strengthen its own synapse while promoting the elimination of synapses formed by other neurons.

The molecular mechanism underlying the formulated “rule of activation” remains unclear. Fig. 19-80 illustrates one possible hypothesis, and another explanation is mentioned in the figure legend. Be that as it may, evidence indicates that this rule holds true for many diverse systems, and we will now examine it in relation to interneuronal synapses.

19.8.7. Synchronously Active Axon Terminals Form Mutually Supportive Synapses [68]

One consequence of the “rule of activation” can be observed in the rat submandibular ganglion, where by the time of birth, each neuron is innervated by axons from approximately five presynaptic neurons located in the Brainstem. By the end of the first postnatal month, through competitive synapse elimination, each ganglion neuron is innervated by only a single such axon. By this time, however, the surviving axon has formed many new terminal branches that establish synapses on other regions of the same cell, so the total number of synapses actually exceeds the initial count (Fig. 19-81). Obviously, Branches of the same axon must share a common property distinguishing them from branches of another axon of the same type: all of them will be excited simultaneously. Yet according to the “rule of activation,” closely spaced endings activated simultaneously mutually reinforce the synapses they form, whereas endings excited independently compete with one another.

19.8.8. The Number of Surviving Synapses Depends on the Number of Dendrites on the Postsynaptic Neuron [69]

Since the competition of synapses for survival is determined in part by the distance between them, the final outcome depends on the Morphology of the postsynaptic cell. Submandibular ganglion neurons are structurally atypical—they lack dendrites; consequently, synaptic competition occurs in close proximity on the cell body, resulting in the preservation of synapses formed by only a single axon. Most other neurons possess numerous dendrites, allowing them to continue receiving signals from diverse sources even in adulthood, which is essential for their integrative function. The role of dendrites in regulating synapse elimination is clearly demonstrated in the rabbit ciliary ganglion, where some neurons have many dendrites while others have very few or none at all (Fig. 19-82). At birth, all neurons are innervated by an equal number of presynaptic axons—roughly four or five. In the adult organism, however, cells lacking dendrites receive signals from only a single axon, whereas the number of axons innervating other cells increases in direct proportion to the number of dendrites. Nevertheless, synapses on a single dendrite are usually formed by endings of the same axon. Therefore, each dendrite presumably represents a separate and independent territory, such that synapses on one dendrite do not compete with those on other dendrites. Much like in skeletal muscle, competition is local and depends on asynchrony of excitation.

Fig. 19-82. Relationship between the number of primary dendrites and the number of axons forming synapses on individual cells in the rabbit ciliary ganglion during development. At birth, the mean number of input synapses is independent of the dendritic count. In adults, the average number of synapses surviving the competitive elimination period is proportional to the number of dendrites. On the right are individual ganglion cells illustrating this relationship. (After D. Purves, R.I. Hume, J. Neurosci. 1: 441-452, 1981, and R.I. Hume, D. Purves, Nature, 293, 469-471, 1981.)

Perhaps the deepest Significance of the activity-dependence of synaptic competition is revealed in instances where external sensory stimuli control the “tuning” of anatomical connections between neurons. This is especially evident in studies of vertebrate visual system development. Relevant findings regarding mammals will be discussed below.

19.8.9. Connections in the Visual System of Infant Mammals Are Modulated by Sensory Experience [70]

At birth, the mammalian visual system remains immature. The first few years (in humans) or months (in cats or monkeys) constitute a distinct sensitive (critical) period during which neural circuitry undergoes fine-tuning, and a lack of normal visual experience during this window can lead to severe and irreversible consequences. A classic example of this is the so-called “lazy eye” resulting from childhood strabismus. Children suffering from strabismus often adapt by using only one eye because the other constantly turns inward, rarely yielding a sharply focused image on its retina. If the strabismus is corrected in time and the child learns to use both eyes, normal visual function will subsequently develop. If left uncorrected in early childhood, however, the disused eye almost entirely and permanently loses its visual capability, rendering lenses useless—a condition known as amblyopia. The eye itself remains structurally normal: the defect resides in the Brain. Before explaining The Nature of this defect, we must examine certain Anatomical Features of the adult mammalian visual system.

19.8.10. Active Synapses Tend to Oust Inactive Ones in the Mammalian Visual System [71]

In mammals such as humans and cats, the visual fields of the two eyes largely overlap, and visual signals from them are combined in the brain to provide binocular stereoscopic Vision. This is made possible because axons relaying signals from equivalent areas of the two retinas form synapses in the same regions of the brain (Fig. 19-83). The primary visual cortex of each cerebral hemisphere contains two orderly “maps” (projections) of the contralateral half of the visual field—one from the left eye and one from the right. These two projections, however, do not overlap perfectly: inputs from the two eyes are segregated into narrow, alternating stripes known as ocular dominance columns. This pattern is schematically illustrated in Fig. 19-83 and can be demonstrated by injecting radioactive Amino Acids into one eye. The labeled molecules are taken up by retinal neurons and transported along nerve cell axons into the visual cortex, somehow traversing synapses within the “Relay stations,” the lateral geniculate nuclei. For instance, autoradiographs of visual cortex slices from an adult monkey clearly show that labeled bands about $0.5\text{ mm}$ wide, receiving information from the “labeled” eye, alternate with unlabeled bands of the same width receiving input signals from the “unlabeled” eye (Fig. 19-84).

Fig. 19-83. Diagram of the main human visual pathways. Input signals from the right and left eyes are distributed in such a way that information about the same regions of the visual field is projected onto the same areas of the brain. Note that all information from the left half of each retina (i.e., from the right half of the visual field) goes to the left hemisphere of the brain, whereas information from the right half of both retinas goes to the right hemisphere.

Fig. 19-84. Ocular dominance columns in the visual cortex of a normal monkey. Radioactive Proline was injected into one eye, and the animal survived for 10 days, during which the radioactive tracer was transported to the cortical areas receiving input from that eye. Tangential sections of the cortex were then prepared to obtain an autoradiograph. When examined under dark-field illumination, the grains overlying the radioactive region appear light against a dark Background. The image shown here is a montage of microphotographs from several serial sections taken through the Cerebral Cortex at different depths. The ocular dominance columns associated with the labeled eye (light stripes) have the same width as those associated with the unlabeled eye (dark stripes). [D. H. Hubel, T. N. Wiesel, S. Le Vay, Philos. Trans. R. Soc. (Biol.), 278, 377-409, 1977.]

However, during development, when visual connections are just beginning to form, no ocular dominance columns can be discerned: the projections of the two retinas completely overlap. Only later (typically in the first weeks of life) do these projections segregate into the alternating stripes characteristic of the adult organism as a result of the competitive elimination of axon terminals. This wiring process appears to be governed by a "co-activation rule": axons transmitting signals from neighboring points on the retina of one eye are usually activated synchronously with one another, but asynchronously with axons carrying signals from the other eye. Axons that fire synchronously reinforce and stabilize the synapses they form on a given cortical cell while displacing synapses from other axons. The segregation of projections into alternating stripes can be halted by artificially stimulating both optic nerves to force axons carrying information from the left eye to fire in strict synchrony with axons from the right eye; alternatively, electrical activity can be suppressed with the same result by injecting tetrodotoxin (which blocks Voltage-Gated Sodium Channels) into both eyes.

The most striking functional consequences are observed when one eye is selectively kept closed during the critical period, thereby depriving it of visual stimulation. When the eye is later opened, the animal behaves as if that eye were totally or partially blind. Autoradiographs reveal that the ocular dominance columns corresponding to the closed eye have sharply narrowed, while those corresponding to the normal eye have expanded to occupy the vacated territory (Fig. 19-85). Consistent with the general rule, synapses formed by inactive axons have vanished, whereas active axons have strengthened and proliferated their synapses. Through this process, cortical territory is allocated to functional inputs and is not wasted on useless axons. Once the critical period has elapsed, this effect becomes irreversible. Thus, the extent and complexity of the cortical apparatus—the number of neurons and synapses by which the adult organism processes sensory information—depend on the sensory stimulation received early in life.

Fig. 19-85. Ocular dominance columns in the cortex of a monkey whose one eye was sutured shut during the critical period. Radioactive proline was injected into the opposite eye, followed by autoradiography (see legend to Fig. 19-84). The ocular dominance columns associated with the visually deprived eye (dark stripes) appear narrowed, whereas those associated with the other eye are expanded. Injecting the tracer into the previously closed eye yields the reverse pattern: narrow light stripes alternating with broad dark ones. [D. H. Hubel, T. N. Wiesel, S. Le Vay, Philos. Trans. R. Soc. (Biol.), 278, 377-409, 1977.]

19.8.11. Formation of converging connections from both eyes requires synchronous binocular stimulation [70, 71, 72]

Early visual experience is also crucial for establishing the neural connections that underlie binocular vision. For example, some children with uncorrected strabismus still use both eyes, but alternately rather than simultaneously. In such cases, both eyes retain visual function, but depth perception (stereoscopic vision) fails to develop. As recordings of electrical activity from single brain cells demonstrate, this phenomenon can likewise be explained by The Influence of neural activity on The Fate of synaptic connections.

Stereoscopic vision relies on binocularly driven neurons—that is, neurons that respond to convergent synaptic inputs from both eyes. Such neurons can be identified in experimental animals by inserting a microelectrode into the visual cortex and recording the responses of individual cells to stimulation of either eye. These cells are located in specific layers of the visual cortex, situated above and below the layer containing "monocular" neurons that form well-defined ocular dominance columns. Normally, an animal possesses an Abundance of "binocular" neurons. However, an animal deprived of synchronous binocular stimulation during the critical period (due to severe strabismus or alternating daily eye patching) nearly lacks such cells altogether. Evidently, binocular neurons maintain their inputs from both eyes only if stimulation of both eyes occurs synchronously. In the absence of synchrony, axons carrying information from one eye compete with axons converging on the same neuron from the other eye; as a result, each neuron eventually retains inputs from only one eye, and the capacity for stereoscopic vision is lost.

19.8.12. The role of the "co-activation rule" in shaping neural connectivity based on individual experience [73]

The development of binocular vision illustrates a broader organizational principle: synchronous activation gives rise to convergent connections. This principle, derived from the "co-activation rule," helps explain how the brain develops neurons that respond specifically to complex combinations of sensory features generated by objects in our environment. For instance, the primate brain contains neurons that appear to be specifically activated by the sight of a particular face. In other words, this clarifies how individual experience can "tune" the brain so that its Structure and function mirror the actual relationships existing among various phenomena in the outside world. In this sense, the rule governing the formation and elimination of synapses during early life forms the foundation of early Learning and Memory.

It has already been suggested in this chapter that memory relies on the modulation of synaptic transmission driven by enduring chemical changes at synapses when Neurotransmitters bind to specific receptor subtypes. Does this mechanism bear any relation to the developmental changes in synaptic connectivity described above? Apparently, at least in some instances, chemical and structural modifications at synapses are intimately linked. For example, when Aplysia undergoes long-term habituation or sensitization (Section 19.5.3) in response to repeated stimulation over several days, chemical alterations at the synapses are accompanied by changes in the size of presynaptic structures. It is also worth noting that in frogs, ocular dominance columns are altered by the administration of NMDA receptor agonists or antagonists, which are likewise believed to participate in the formation of memory traces in the hippocampus (Section 19.5.6).

Beyond the evidence discussed above, everything else remains speculative; the Mechanisms of Memory and of synapse formation and elimination are still largely poorly understood. It is clear, however, that these issues stand among the central problems of neurobiology, and their resolution will likely broaden and unify our understanding of brain function across multiple levels.

Conclusions

Synapses first form early in development, but the initial pattern of connections undergoes extensive remodeling through the elimination of old synapses and the creation of new ones. The formation of a neuromuscular junction between a motoneuron and a muscle fiber triggers changes in both cells, accompanied by the accumulation of specific molecules in the intervening basal lamina. This specialized basal lamina of the neuromuscular junction persists even after the destruction of both the motor axon and the muscle cell, and it guides the restoration of synaptic components in both the muscle fiber and the axon terminal during regeneration.

In mammals at birth, each muscle fiber typically receives multiple synaptic inputs, all but one of which are subsequently eliminated through competition. Activation of a muscle fiber via one synaptic contact promotes the elimination of neighboring synapses that happen to be inactive at that moment. When pre- and postsynaptic activity coincide in time, the synapse is stabilized and strengthened. This so-called "co-activation rule," whose molecular mechanism remains elusive, appears to account for synapse formation and elimination across many regions of the developing nervous system. In particular, this rule helps explain how the brain's synaptic circuitry is "tuned" in accordance with the individual animal's experience.

References

General

The Brain, Sci. Am., 241(3), 1979.

Cooke I., Lopkin M., eds. Cellular Neurophysiology: A Source Book. New York: Holt, Rinehart and Winston, 1972.

Hille B. Ionic Channels of Excitable Membranes. Sunderland, MA: Sinauer, 1984.

Kandel E. R., Schwartz J. H. Principles of Neural Science, 2nd ed. New York: Elsevier, 1985.

Kuffler S. W., Nicholls J. G., Martin A. R. From Neuron to Brain, 2nd ed. Sunderland, MA: Sinauer, 1984.

Molecular Neurobiology. Cold Spring Harbor Symp. Quant. Biol., 48, 1983.

Patterson P. H., Purves D. Readings in Developmental Neurobiology. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory, 1982.

Purves J., Lichtman J. M. Principles of Neural Development. Sunderland, MA: Sinauer, 1985.

Ramon y Cajal S. Histologie du Systeme Nerveux de l'Homme et des Vertebres. Paris: Maloine, 1909-1911. (Reprinted, Madrid: Consejo Superior de Investigaciones Cientificas, Institute Ramon y Cajal, 1972.)

Cited

1. Bullock T. H., Orkand R., Grinnell A. Introduction to Nervous Systems, pp. 6 and 393-496. San Francisco: Freeman, 1977.

Nauta W. J. H., Feirtag M. Fundamental Neuroanatomy. New York: Freeman / Scientific American Library, 1986.

2. Ramon y Cajal S. Recollections of My Life (E. H. Craigie, trans.). In: Memoirs of the American Philosophical Society. Vol. 8. Philadelphia, 1937. Reprinted, New York: Garland, 1988.

Stevens C. F. The neuron. Sci. Am., 241(3), 48-59, 1979.

3. Hodgkin A. L. The Conduction of the Nervous Impulse. Liverpool, U. K.: Liverpool University Press, 1964.

4. Katz B. Nerve, Muscle and Synapse. New York: McGraw-Hill, 1966.

5. Bunge M. B. The axonal cytoskeleton: its role in generating and maintaining cell form. Trends Neurosci., 9, 477-482, 1986.

Grafstein B., Forman D. S. Intracellular Transport in neurons. Physiol. Rev., 60, 1167-1283, 1980.

Peters A., Palay S. L., Webster H., de F. The Fine Structure of the Nervous System. Philadelphia: Saunders, 1976.

Schwartz J. H. The Transport of substances in Nerve Cells. Sci. Am., 242(4), 152-171, 1980.

6. Jones E. G. Pathways to progress—the rise of modern neuroanatomical techniques. Trends Neurosci., 9, 502-505, 1986.

LaVail J. H., LaVail M. M. Retrograde axonal transport in the Central Nervous System. Science, 176, 1416-1417, 1972.

Schnapp B. J., Vale R. D., Sheetz M. P., Reese T. S. Single microtubules from squid axoplasm support bidirectional movement of Organelles. Cell, 40, 455-462, 1985.

Vale R. D. Intracellular transport using microtubule-based motors. Annu. Rev. Cell Biol., 3, 347-378, 1987.

7. Goldstein G. W., Betz A. L. The Blood-brain barrier. Sci. Am., 255(3), 74-83, 1986.

Kuffler S. W., Nicholls J. G., Martin A. R. From Neuron to Brain, 2nd ed., pp. 323-375. Sunderland, MA: Sinauer, 1984.

Perry V. H., Gordon S. Macrophages and microglia in the nervous system. Trends Neurosci., 11, 273-277, 1988.

8. Hille B. Ionic Channels of Excitable Membranes, pp. 21-75. Sunderland, MA: Sinauer, 1984.

Kandel E. R., Schwartz J. H. Principles of Neural Science, 2nd ed., pp. 49 - 86. New York: Elsevier, 1985.

Kuffler S. W., Nicholls J. G., Martin A. R. From Neuron to Brain, 2nd ed., pp. 97-206. Sunderland, MA: Sinauer, 1984.

9. Hodgkin A. L., Rushton W. A. H. The electrical constants of a crustacean nerve fibre. Proc. R. Soc. Lond. (Biol.)., 133, 444-479, 1946.

Roberts A., Bush В. M. H., eds. Neurones Without Impulses. Cambridge, U.K.: Cambridge University Press, 1981.

10. Baker P.P., Hodgkin A. L., Shaw T. The effects of changes in internal ionic concentrations on the electrical properties of perfused giant axons. J. Physiol., 164, 355-374, 1962.

Chiu S. Y., Ritchie J. M., Rogart R. В., Stagg D. A quantitative desciption of membrane currents in rabbit myelinated nerve. J. Physiol., 292, 149-166, 1979.

Hodgkin A. L. Chance and design in electrophysiology: an informal account of certain experiments on nerve carried out between 1934 and 1952. Hodgkin A. L., Huxley A. F. Currents carried by sodium and potassium ions through the membrane of the giant axon of Loligo. J. Physiol., 116, 449-472, 1952.

Hodgkin A.L., Huxley A. P., Katz B. Measurement of currnet-voltage relations in the membrane of the giant axon of Loligo. J. Physiol., 116, 424448, 1952.

Hodgkin A.L., Katz B. The cfflect of sodium ions on the electrical activity of the giant axon of the squid. J. Physiol., 108, 37-77, 1949.

11. Hodgkin A. L., Huxley A. F. A quantitative description of membrane current and its application to conduction and excitation in nerve. J. Physiol., 117, 500 544, 1952.

12. Bray G.M., Rasminsky M., Aquayo A.J. Interactions between axons and their sheath cells. Annu, Rev. Neurosci., 4, 127-162, 1981. French-Constant C., Raff M. C. The oligodendrocyte-type 2 astrocyte cell Lineage is specialized for myelination. Nature, 323, 335-338, 1986, Morell P., Norton W.T. Myelin. Sci. Am., 242(5), 88-118, 1980.

13. Brown D.A. Synaptic mechanisms. Trends Neuroci., 9, 468-470, 1986.

Katz B. Nerve, Muscle and Synapse, pp. 97-158. New York: McGraw-Hill, 1966.

Kuffler S. W., Nicholls J. G., Martin A. R. From Neuron to Brain, 2nd ed., pp. 207-320. Sunderland, MA: Sinauer, 1984.

14. Dale H. H., Feldberg W., Vogt M. Release of acetylcholine at voluntary motor nerve endings. J. Physiol., 86, 353-380, 1936.

Fatt P., Katz B. An Analysis of the end-plate potential recorded with an intracellular electrode. J. Physiol., 115, 320-370, 1951.

Feldberg W. The early history of synaptic and neuromuscular transmission by acetylcholine: reminiscences of an eyewitness. In The Pursuit of Nature (A. L. Hodgkin, et al.), pp- 65-83. Cambridge, U.K.: Cambridge University Press, 1977.

15. Hill B. Ionic Channels of Excitable Membranes, pp. 76-98. Sunderland, MA: Sinauer, 1984.

Katz В., Miledi R. The timing of calcium action during neuromuscular transmission. J. Physiol., 189, 535-544, 1967.

Llinas R. Calcium in synaptic transmission. Sci. Am., 247(4), 56-65, 1982.

Miller R.J. Calcium signalling in neurons. Trends Neurosci., 11, 415-419, 1988.

16. Couteaux R., Pecot-Dechavassine M. Vesicules synaptiques et poches au niveau des zones actives de la jonction neuromusculaire. Comptes Rendus Acad. Sci. (Paris) D, 271, 2346-2349, 1970.

Heuser J. E., et al. Synaptic vesicle exocytosis captured by quick freezing and correlated with quantal transmitter release. J. Cell Biol. 81:275-300, 1979.

Hauser J. E., Reese T. S. Structural changes after transmitter release. Trends Neurosci., 11, 458-464, 1988.

17. Del Castillo J., Katz B. Quantal Components of the endplate potential. J. Physiol, 124, 560 573, 1954.

Fatt P., Katz B. Spontaneous subthreshold activity at motor nerve endings. J. Physiol., 117, 109-128, 1952.

18. Hille B. Ionic Channels of Excitable Membranes, pp. 117-147. Sunderland, MA: Sinauer, 1984.

19. Lester H. A. The response to acetylcholine. Sci. Am., 236(2), 106-118, 1977.

Sakmann В., Bormann J., Hamill O. P. Ion transport by single receptor channels. Cold Spring Harbor Symp. Quant. Biol., 48, 247-257, 1983.

20. Massoulie J.. Bon S. The molecular forms of cholinesterase and acetylcholinesterase in vertebrates. Annu. Rev. Neurosci., 5, 57-106, 1982. Taylor P., Schumacher M., MacPhee-Quigley K., Friedmann T, Taylor S. The structure of acetylcholinesterase: relationship to its function and cellular disposition. Trends Neurosci., 10, 93-95, 1987.

21. Barnard E. A., Darlison M.G., Seeburg P. Molecular biology of the GABA-A receptor: the receptor/channel superfamily. Trends Neurosci., 10, 502-509, 1987.

Grenningloh G., et al. The strychnine-binding subunit of the Glycine receptor shows Homology with nicotinic acetylcholine receptors. Nature, 328, 215-220, 1987.

Hille B. Ionic Channels of Excitable Membranes, pp. 371-383. Sunderland, MA: Sinauer, 1984. (Evolution of channels.)

22. Gottlieb D.I. GABAergic neurons. Sci. Am., 285(2), 38-45, 1988.

Mayer M. L., Westbrook G. L. The physiology of excitatory amino acids in the vertebrate central nervous system. Prog. Neurobiol., 28, 197-276, 1987.

Sneddon P., Westfall D. P. Pharmacological evidence that adenosine triphosphate and noradrenaline are cotransmitters in the quinea pig vas deferens. J. Physiol., 347, 561-580, 1984.

23. Snyder S.H. Drug and neurotransmitter receptors in the brain. Science, 224, 22-31, 1984.

Nathanson N. M. Molecular properties of the muscarinic Acetylcholine Receptor. Annu. Rev. Neurosci., 10, 195-236, 1987.

24. Bormann J. Electrophysiology of GABA-A and GABA-B receptor subtypes. Trend Neurosci., 11, 112-116, 1988.

Snyder S.H. Drug and the Brain. New York: W. H. Freeman/Scientific American Books, 1987.

Tollman J. F., Gallager D. W. The GABAergic system: a locus of benzodiazepine action. Annu. Rev. Neurosci., 8, 21-44, 1985.

25. Kuffler S. W., Nicholls J. G., Martin А. Я From Neuron to Brain, 2nd ed., pp. 407-430. Sunderland, MA: Sinauer, 1984.

26. Barren J. N. Motoneuron dendrites: role in synaptic integration. Fed. Proc., 34, 1398-1407, 1975.

27. Coombs J. S., Curtis D. R., Eccles J. C. The generation of impulses in motoneurones. J. Physiol., 139, 232-249, 1957.

Fuortes M. G. F., Frank K., Becker M. C. Steps in The production of motoneuron spikes. J. Gen. Physiol., 40, 735-752, 1957.

28. Connor J. A., Stevens C. F. Prediction of repetitive firing behaviour from Voltage Clamp data on an isolated neurone soma. J. Physiol., 213, 3153, 1971.

Hille B. Ionic Channels of Excitable Membranes, pp. 99-116. Sunderland, MA: Sinauer, 1984.

Rogawski M. A. The A-current: how ubiquitous a feature of excitable cells is it? Trends Neurosci., 8, 214-219, 1985.

29. Meech R. W. Calcium-dependent potassium activation in nervous Tissues. Annu. Rev. Biophys. Bioeng., 7, 1-18, 1978.

Tsien R. W., Lipscombe D., Madison D. V., Bley K.R., Fox A. P. Multiple types of neuronal calcium channels and their selective modulation.

Trends Neurosci., 11, 431-438, 1988.

30. Bullock T. H., Horridge G. A. Structure and Function in the Nervous System of Invertebrates, pp. 38-124. San Franciso: Freeman, 1965.

Llinas R., Sugimori M. Electrophysiological properties of in vitro Purkinje cell dendrites in mammalian cerebellar slices. J. Physiol., 305, 197-213, 1980.

Shepherd G.M. Microcircuits in the nervous system. Sci. Am., 238(2), 92-103, 1978.

31. Breitwieser G. E., Szabo G. Uncoupling of cardiac muscarinic and betaadrenergic receptors from ion channels by a guanine nucleotide analogue. Nature, 317, 538-540, 1985.

Levitan I. B. Modulation of ion channels in neurons and other cells. Annu. Rev. Neurosci., 11, 119-136, 1988.

Nairn A. C., Hemmings H. C., Greengard P. Protein Kinases in the brain. Annu. Rev. Biochem., 54, 931 -976, 1985.

Pfaffinger P. J., Martin J. M., Hunter D. D., Nathanson N. M., Hille B. GTP-binding Proteins couple cardiac muscarinic receptors to а К channel. Nature, 317, 536-538, 1985.

32. Iversen L.L. The Chemistry of the brain. Sci. Am., 241(3), 118-129, 1979.

Snyder S.H. Drugs and the Brain. New York: W. H. Freeman/Scientific American Library, 1988.

33. Bloom F.E., Neuropeptides. Sci. Am., 245(4), 148-168, 1981.

Hokfelt Т., Johanson O., Goldstein M. Chemical Anatomy of the brain. Science, 225, 1326-1334, 1984.

Jan V. N., Bowers C. W., Branton D., Evans L., L. Y. Peptides in neuronal function: studies using frog autonomic ganglia. Cold Spring Harbor Symp. Quant. Biol., 48, 363-374, 1983.

Scheller R. H., Axel R. How genes control an innate behavior. Sci. Am., 250(3), 44- 52, 1984. (Neuropeptides in Aplysia).

34. Parley J., Alkon D. Cellular mechanisms of learning, memory, and information storage. Annu. Rev. Psychol., 36, 419 494, 1985.

Kandel E.R. Small systems of neurons. Sci. Am., 241(3), 60-70, 1979.

Morris R. G. M., Kandel E. R., Squire L. R., eds. Learning and Memory. Trends Neurosci., 11, 125-181, 1988.

35. Greenberg S. M., Castellucci V. F., Bayley H., Schwartz J. H. A molecular mechanism for long-term sensitization in Aplysia. Nature, 329, 6265, 1987.

Montarolo P. G., et al. A critical period for macromolecular synthesis in long-term heterosynaptic facilitation in Aplysia. Science, 234, 1249-1254, 1986.

Schwartz J. H., Greenberg S. M. Molecular mechanisms for memory: second-messenger induced modifications of protein kinases in nerve cells. Annu. Rev. Neurosci., 10, 459-476, 1987.

Siegelbaum S. A., Camardo J. S., Kandel E. R. Serotonin and cyclic AMP close single K+ channels in Aplysia sensory neurons. Nature, 299, 413417, 1982.

36. Dudai Y. Neurogenetic dissection of learning and short-term memory in Drosophila. Annu. Rev. Neurosci., 11, 537-563, 1988.

Kandel E. R., et al. Classical conditioning and sensitization share aspects of the same molecular cascade in Aplysia. Cold Spring Harbor Symp. Quant. Biol., 48, 821-830, 1983.

37. Bliss T. V. P., Lomo T. Long-lasting potentiation of synaptic transmission in the dentate area of the anaesthetized rabbit following stimulation of the perforant path. J. Physiol., 232, 331 356, 1973.

Collingridge G. L., Bliss T.V.P. NMDA receptors and their role in long-term potentiation. Trends Neurosci., 10, 288-293, 1987

Cotman C. W., Monaghan D. Т., Ganong A.H. Excitatory amino acid neurotransmission: NMDA receptors and Hebb-type synaptic plasticity. Annu. Rev. Neurosci, 11, 61-80, 1988.

Lisman J.E., Goldring M.A. Feasibility of long-term storage of graded information by the Ca²⁺ /calmodulin-dependent protein kinase molecules of the postsynaptic density. Proc. Natl. Acad. Sci. USA, 85, 5320-5324, 1988.

Mishkin M., Appenzeller T. The anatomy of memory. Sci. Am., 256(6), 80-89, 1987.

Morris R. G., Anderson E., Lynch G., Baudry M. Selective impairment of learning and blockade of long-term potentiation by an N-methyl-D- aspartate receptor antagonist, AP5. Nature, 319, 774-776, 1986.

38. Barlow H. В., Mollon J. D., eds. The Senses. Cambridge, U. K.: Cambridge University Press, 1982.

Schmidt R. F., ed. Fundamentals of Sensory Physiology. New York: Springer, 1978.

Shepherd G. Neurobiology, 2nd ed., pp. 205-353. New York: Oxford University Press, 1988.

39. Katz B. Depolarization of sensory terminals and the initiation of impulses in the muscle spindle. J. Physiol., 111, 261-282, 1950.

40. Hudspeth A. J. The Cellular Basis of Hearing: the biophysics of Hair cells. Science, 230, 745-752, 1985.

Roberts W. M., Howard J., Hudspeth A. J. Hair cells: Transduction, tuning, and transmission in the Inner ear. Annu. Rev. Cell Biol., 4, 63-92, 1988.

Von Bekesy G. The ear. Sci. Am., 197(2), 66—78, 1957.

41. Corey D.P., Hudspeth A.J. Compliance of the hair bundle associated with medianoelectrical transduction channels in the bullfrog's saccular hair cell. Neuron, 1, 189-199, 1988.

Pickles J. О. Recent advances in cochlear physiology. Prog. Neurobiol., 24, 1 -42, 1985.

42. Barlow H.B., Mollon J.D., eds. The Senses, pp. 102-164. Cambridge, U.K.: Cambridge University Press, 1982.

43. Baylor D.A., Lamb T.D., Yau K. W. Responses of retinal rods to single photons. J. Physiol., 288, 613-634, 1979.

Schnapf J.L., Baylor D. A. How photoreceptor cells respond to light. Sci. Am, 256(4), 40-47, 1987.

44. Stryer L. The molecules of visual excitation. Sci. Am., 257(1), 32-40, 1987.

45. Fesenko E. E., Kolesnikov S.S., Lyubarsky A. L. Induction by cyclic GMP of cationic conductance in plasma membrane of retinal rod outer segment. Nature, 313, 310-313, 1985.

Stryer L. The cyclic GMP cascade of vision. Annu. Rev. Neurosci., 9, 87-119, 1986.

46. Koch K. W., Stryer L. Highly cooperative feedback control of retinal rod guanylate cyclase by Calcium Ions. Nature, 334, 64-66, 1988.

Matthews H. R., Murphy R.L.W., FainG.L., Lamb T.D. Photoreceptor hight adaptation is mediated by cytoplasmic calcium concentration. Nature, 334, 67-69, 1988.

Nakatani K., Yau K. W. Calcium and light adaptation in retinal rods and cones, Nature, 334, 67-71, 1988.

47. Hubel D.H. Eye, Brain, and Vision. New York: W. H. Freeman / Scientific American Library, 1988.

Kuffler S. W., Nicholls J. G., Martin A. R. From Neuron to Brain, 2nd ed., pp. 19-96. Sunderland, MA: Sinauer, 1984.

Masland R. H. The functional architecture of the retina. Sci. Am., 255(6), 102-111, 1986.

48. Cowan W.M. The Development of the brain. Sci. Am., 241(3), 106-117, 1979.

Hopkins W. G., Brown M. C. Development of Nerve Cells and their Connections. Cambridge, U.K.: Cambridge University Press, 1984.

Parnavelas J. G., Stern C. D., Stirling R. V., eds. The Making of the Nervous System. Oxford, U.K.: Oxford University Press, 1988.

Purves D., Lichtman J. W. Principles of Neural Development. Sunderland, MA: Sinauer, 1985.

49. Alvarez-Buylla A., Nottebohm F. Migration of young neurons in aduit avian brain. Nature, 335, 353-354, 1988.

Le Douarin N. M., Smith J. Development of the Peripheral Nervous System from the neural crest. Annu. Rev. Cell Biol., 4, 375-404, 1988.

Stent G. S., Weisblat D. A. Cell lineage in the development of invertebrate nervous system. Annu. Rev. Neurosci., 8, 45-70, 1985.

Williams R. W., Herrup K. The control of neuron number. Annu. Rev. Neurosci., 11, 423-453, 1988.

50. Hollyday M., Hamburger V. An autoradiographic Study of the Formation of the lateral motor Column in the chick embryo. Brain Res., 132, 197208, 1977.

Rakic P., Mode of cell migration to the superficial layers of the fetal monkey neocortex. J. Сотр. Neurol., 145, 61-84, 1972.

51. Caviness V.S. Neocortical histogenesis in normal and reeler mice: a developmental study based on [3H] thymidine autoradiography. Dev. Brain Res., 4, 293-302, 1982.

McConnell S. K. Development and decision-making in the mammalian cerebral cortex. Brain Res. Rev., 13, 1-23, 1988.

Rakic P. Specification of cerebral cortical areas. Science, 241, 170-176, 1988.

52. Bray D., Hollenbeck P. J. Growth cone motility and guidance. Annu. Rev. Cell Biol., 4, 43-61, 1988. Harrison R.G. The outgrowth of the nerve fiber as a mode of protoplasmic movement. J. Exp. Zool., 9, 787-846, 1910.

Yamada K. M., Spooner B. S., Wessells N. K. Ultrastructure and function of growth cones and axons of cultured nerve cells. J. Cell Biol., 49, 614635, 1971.

53. Bamburg J. B. The axonal cytoskeleton: stationary or moving matrix? Thrends Neurosci., 11, 248-249, 1988.

Bamburg J. R., Bray D., Chapman K. Assembly of microtubules at the tip of growing axons. Nature, 321, 788-790, 1986.

Bray D. Surface movements during the growth of single explanted neurons. Proc. Natl. Acad. Sci. USA 65: 905-910, 1970.

Letourneau P. C., Ressler A. N. Inhibition of neurite initiation and growth by taxol. J. Cell Biol., 98, 1355-1362, 1984.

54. Davies A. M. MOLECULAR AND CELLULAR aspects of patterning sensory neurone connections in the vertebrate nervous system. Development, 101, 185-208, 1987.

Kater S. В., Mattson M. P., Cohan C., Connor J. Calcium Regulation of the neuronal growth cone. Trends Neurosci., 11, 315-321, 1988.

Letourneau P. C. Cell-to-substratum adhesion and guidance of axonal elongation. Dev. Biol., 44, 92-101, 1975.

Lumsden A. G.S., Davies A. M. Earliest sensory nerve fibres are guided to peripheral targets by attractants other than Nerve Growth Factor. Nature, 306, 786-788, 1983.

Patel N., Poo M. M. Orientation of neurite growth by extracellular electric fields. J. Neurosci., 2, 483-496, 1984.

55. Bentley D., Caudy M. Navigational substrates for peripheral pioneer growth cones: limb-axis polarity cues, limb-segment boundaries, and guidepost neurons. Cold Spring Harbor Symp. Quant. Biol., 48, 573-585, 1983.

Berlot J., Goodman C. S. Guidance of peripheral pioneer neurons in the grasshopper: adhesive hierarchy of epithelial and neuronal surfaces. Science, 223, 493-496, 1984.

Goodman C. S., Bastiani M. J. How embryonic nerve cells recognize one another. Sci. Am., 251(6), 58-66, 1984.

56. Bixby J. L., Pratt R. S., Lilien J., Reichardt L. F. Neurite outgrowth on muscle cell surfaces involves extracellular matrix receptors as well as Ca²+-dependent and independent Cell Adhesion molecules. Proc. Natl. Acad. Sci. USA, 84, 2555-2559, 1987.

Chang S., Rathjen F. G., Raper J. A. Extension of neurites on axons is impaired by antibodies against specific neural cell adhesion molecules. J. Cell Biol., 104, 355-362, 1987.

Jessell T. M. Adhesion molecules and the hierarchy of neural development. Neuron, 1, 3 13, 1988.

Sanes J. R., Schaehner M., Covault J. Expression of several adhesive macromolecules (N-CAM, L1, Jl, NILE, uvomorulin, Laminin, Fibronectin, and a heparan sulfate protcoglycan) in embryonic, adult, and denervated adult skeletal muscle. J. Cell Biol., 102, 420 431, 1986.

Tomaselli K.J., el al. N-cadherin and Integrins: two receptor systems that mediate neuronal process outgrowth on astrocyte surfaces. Neuron, 1; 3343, 1988.

57. Lance-Jones C., Landmesser L. Motoneurone projection patterns in the chick hindlimb following early partial reversals of the Spinal Cord. J. Physiol., 302, 581-602, 1980. Landmesser L. The development of specific motor pathways in the chick embryo. Trends Neurosci., 7, 336-339, 1984.

Sperrv R. W. Chemoaffinity in the orderly growth of nerve fiber patterns and connections. Proc. Natl. Acad. Sci. USA, 50, 703-710, 1963.

Udin S. В., Fawcett J. W. Formation of topographic maps. Annu. Rev. Neurosci., 11, 289-327, 1988.

58. Campenot R. B. Local control of neurite development by nerve growth factor. Proc. Natl. Acad. Sci. USA, 74, 4516-4519, 1977.

Davies A., Lumsden A. Relation of target encounter and neuronal death to nerve growth factor responsiveness in the developing mouse trigeminal ganglion. J. Сотр. Neurol., 223, 124-137, 1984.

Greene L. A. The importance of both early and delayed responses in the biologien actions of nerve growth factor. Trends Neurosci., 7, 91 94, 1984.

Levi-Montalcini R., Calissano P. The nerve growth factor. Sci. Am., 240(6), 68-77, 1979.

59. Cowan W.M., Fawcett J. W., O'Leary D.D.M., Stanfield В. В. Regressive events in neurogenesis. Science, 225, 1258 1265, 1984.

Davies A.M. Role of Neurotrophic Factors in development. Trends Genet., 4, 139-144, 1988.

Hamburger V., Levi-Montalcini R. Proliferation, differentiation and degeneration in the spinal ganglia of the chick embryo under normal and experimental conditons. J. Exp. Zool., 162, 133-160, 1949.

Hamburger V., Vip J. W. Reduction of experimentally induced neuronal death in spinal ganglia of the chick embryo by nerve growth factor. J. Neurosci., 4, 767-774, 1984.

Williams R. W., Herrup K. The control of neuron number. Annu. Rev. Neurosci., 11, 423-453, 1988.

60. Brown M. C., Holland R. L., Hopkins W. G. Motor nerve sprouting. Annu. Rev. Neurosci., 4, 17-42, 1981.

Davies A.M. The survival and growth of embryonic proprioceptive neurons is promoted by a factor present in skeletal muscle. Dev. Biol., 115, 5667, 1986.

Ebendal Т.. Olson L., Seiger A., Hedlund К. О. Nerve growth factors in the rat iris. Nature, 286, 25-28, 1980.

Purves D., Voyvodic J. T. Imaging mammalian nerve cells and their connections over time in living animals. Trends Neurosci., 10, 398 404, 1987.

61. Purves D., Lichtman J.W. Principles of Neural Development. Sunderland, MA: Sinaucr, 1985.

62. Hume R. I., Role L. W., Fischbach G. D. Acetylcholine release from growth cones detected with patches of acetylcholine receptor-rich membranes. Nature, 305, 632-634, 1983.

Jarmillo F., Vtcini S.. Schuetze S. M. Embryonic acetylcholine receptors guarantee spontaneous contractions in rat developing muscle. Nature, 335, 66 68, 1988.

Spitzer N. C. Ion channels in development. Annu. Rev. Neurosci., 2, 363-397, 1979.

Young S. H., Poo M. M. Spontaneous release of transmitter from growth cones of embryonic neurones. Nature, 305, 634-637, 1983.

63. Poo M.M. Mobility and localization of proteins in excitable membranes. Annu. Rev. Neurosci., 8, 369-406, 1985.

Schuetze S. M., Role L. M. Development regulation of nicotinic acetylcholine receptors. Annu. Rev. Neurosci., 10, 403-457, 1987.

64. Burden S. J., Sargent P. В., McMagan U.J. Acetylcholine receptors in regenerating muscle accumulate at original synaptic sites in the absence of the nerve. J. Cell Biol., 82, 412-425, 1979.

Nitkin R. M., el al. Identification of agrin, a synaptic organizing protein from Torpedo electric organ. J. Cell Biol., 105, 2471-2478, 1987.

Sanes J. R., Hall Z. W. Antibodies that bind specifically to synaptic sites on muscle fiber basal lamina. J. Cell Biol., 83, 357 370, 1979.

65. Frank E., Jansen J. K. S., Lomo Т., Westgaard R. H. The interaction between foreign and original motor nerves innervating the soleus muscle of rats. J. Physiol., 247, 725-742, 1975.

Jones R., Vrbova G. Two factors responsible for denervation hypersensitivity. J. Physiol., 236, 517-538, 1974.

Lomo Т., Jansen J. K. S. Requirement for the formation and maintenance of neuromuscular connections. Curr. Top. Dev. Biol., 16, 253-281, 1980. Lomo Т., Rosenthal J. Control of ACh sensitivity by muscle activity in the rat. J. Physiol., 221, 493-513, 1972.

66. Oppenheim R. W. Cell death during neural development. In: Handbook of Physiology, Vol. 1: Neuronal Development, (W. M. Cowan, ed.). Washington, DC: American Physiological Society, 1988.

Pittman R., Oppenheim R. W. Cell death of motoneurons in the chick embryo spinal cord. IV. Evidence that a functional neuromuscular interaction is involved in The regulation of naturally occurring cell death and the stabilization of synapses. J. Сотр. Neurol, 187, 425-446, 1979.

67. Brown M. C., Jansen J. K. S. Van Essen D. Polyneuronal innervation of skeletal muscle in new-born rats and its elimination during maturation. J. Physiol., 261, 387-422, 1976.

Callaway E. M., Soha J. M., Van Essen D. Competition favouring inactive over active motor neurons during synapse elimination. Nature, 328, 422—426, 1987.

O'Brien R.A.D., Ostberg A.J.C., Vrbova G. Protease inhibitors reduce the lose of nerve terminals induced by activity and calcium in developing rat soleus Muscles in vitro. Neurosci., 12, 637-646, 1984.

Ribshester R. R., Taxt T. Motor unit size and synaptic competiton in ral lumbrical muscles reinnervated by active and inactive motor axons. J. Physiol., 344, 89 111, 1983.

68. Lichtman J. W. The reorganization of synaptic connexions in the rat submandibular ganglion during post-natal development. J. Physiol., 273, 155-177, 1977.

Purves D., Lichtman J.W. Principles of Neural Development, pp. 271-328. Sunderland, MA: Sinauer, 1985.

69. Purves D. Modulation of neuronal competiton by postsynaptic geometry in autonomic ganglia. Trends Neurosci., 6, 10-16, 1983.

70. Barlow H. Visual experience and cortical development. Nature, 258, 199-204, 1975. Wiesel T.N. Postnatal development of the visual cortex and the influence of environment. Nature, 299, 583-591, 1982.

71. Fawcett J. W. Retinotopic maps, cell death, and electrical activity in the retinotectal and retinocollicular projections. In: The Making of the Nervous System (J.G. Parnavelas; C.D. Stern; R.V. Stirling, eds.), pp. 395-416. Oxford, U.K.: Oxford University Press, 1988.

Hubel D. H., Wiesel T.N. Ferrier lecture: functional architecture of macaque monkey visual cortex. Proc. R. Soc. Lond. Biol., 198, 1 59, 1977. Hubel D. H., Wiesel T.N., Le Vay S. Plasticity of ocular dominance columns in monkey striate cortex. Philos. Trans. R. Soc. Lond. Biol., 278, 377409, 1977.

Rakic P. Prenatal genesis of connections subserving ocular dominance in the rhesus monkey. Nature, 261, 467-471, 1976.

Stryker M.P., Harris W.A. Binocular impulse blockade prevents the formation of ocular dominance columns in cat visual cortex. J. Neurosci., 6, 2117-2133, 1986.

72. Hubel D. H., Wiesel T.N. Binocular interaction in striate cortex of kittens reared with artificial squint. J. Neurophysiol., 28, 1041-1059, 1965.

Le Vay S.. Wiesel T. N., Hubel D. H. The development of ocular dominance columns in normal and visually deprived monkeys. J. Сотр. Neurol, 191, 1 51, 1980.

73. Baylis G. C., Rolls E. Т., Leonard С. М. Selectivity between faces in the responses of a population of neurons in the cortex in the superior temporal sulcus of the monkey. Brain Res., 342, 91-102, 1985.

Cline H. Т., Debski E.A., Constantine-Paton M. N-methyl-D-aspartate receptor antagonist desegregates eye-specific stripes. Proc. Natl. Acad. Sci. USA, 84, 4342-4345, 1987,

Greenough W. Т., Bailey С. Н. The anatomy of a memory: convergence of results across a diversity of tests. Trends Neurosci., 11, 142-147, 1988. Perrett D.I., Mistlin A.J., Chitty A.J. Visual neurones responsive to faces. Trends Neurosci., 10, 358-364, 1987.