Chapter 2: Development and Cellular Constituents of the Nervous System

Early in the development of the nervous system, a hollow tube of ectodermal neural tissue forms at the embryo's dorsal midline. The cellular elements of the tube appear undifferentiated at first, but they later develop into various types of neurons and supporting glial cells.

Layers of the Neural Tube

The embryonic neural tube has three layers (Fig 2–1): the ventricular zone, later called the ependyma, around the lumen (central canal) of the tube; the intermediate zone, which is formed by the dividing cells of the ventricular zone (including the earliest radial glial cell type) and stretches between the ventricular surface and the outer (pial) layer; and the external marginal zone, which is formed later by processes of the nerve cells in the intermediate zone (Fig 2–1B).

The intermediate zone, or mantle layer, increases in cellularity and becomes gray matter. The nerve cell processes in the marginal zone, as well as other cell processes, become white matter when myelinated.

Differentiation and Migration

The largest neurons, which are mostly motor neurons, differentiate first. Sensory and small neurons, and most of the glial cells, appear later, up to the time of birth. Newly formed neurons may migrate extensively through regions of previously formed neurons. When glial cells appear, they can act as a framework that guides growing neurons to the correct target areas. Because the axonal process of a neuron may begin growing toward its target during migration, nerve processes in the adult brain are often curved rather than straight.

Neurons vary in size and complexity. Motor neurons are usually larger than sensory neurons. Nerve cells with long processes (eg, dorsal root ganglion cells) are larger than those with short processes (Figs 2–2 and 2–3).

Some neurons project from the cerebral cortex to the lower spinal cord, a distance of 4 ft or more in adults; others have very short processes, reaching, for example, only from cell to cell in the cerebral cortex. These small neurons, with short axons that terminate locally, are called interneurons.

Extending from the nerve cell body are usually a number of processes called the axon and dendrites. Most neurons give rise to a single axon (which branches along its course) and to many dendrites (which also divide and subdivide, like the branches of a tree). The receptive part of the neuron is the dendritic zone (see Dendrites section). The conducting (propagating or transmitting) part is the axon, which may have one or more collateral branches. The downstream end of the axon is called the synaptic terminal, or arborization. The neuron's cell body is called the soma, or perikaryon.

Cell Bodies

The cell body is the metabolic and genetic center of a neuron (see Fig 2–3). Although its size varies greatly in different neuron types, the cell body makes up only a small part of the neuron's total volume.

The cell body and dendrites constitute the receptive pole of the neuron. Synapses from other cells or glial processes tend to cover the surface of a cell body (Fig 2–4).

Dendrites

Dendrites are branches of neurons that extend from the cell body; they receive incoming synaptic information and thus, together with the cell body, provide the receptive pole of the neuron. Most neurons have many dendrites (see Figs 2–2, 2–3, and 2–5). Because most dendrites are long and thin, they act as resistors, isolating electrical events, such as postsynaptic potentials, from one another (see Chapter 3). The branching pattern of the dendrites can be complex and determines how the neuron integrates synaptic inputs from various sources. Some dendrites give rise to dendritic spines, which are small mushroom-shaped projections that act as fine dendritic branches and receive synaptic inputs (Fig 2–5). Dendritic spines are currently of great interest to researchers. The shape of a spine regulates the strength of the synaptic signal that it receives. A synapse onto the tip of a spine with a thin "neck" will have a smaller influence than a synapse onto a spine with a thick neck. Dendritic spines are dynamic, and their shape can change. Changes in dendritic spine shape can strengthen synaptic connections so as to contribute to learning and memory. Maladaptive changes in spines may contribute to altered function of the nervous system after injury.

Axons

A single axon or nerve fiber arises from most neurons. The axon is a cylindrical tube of cytoplasm covered by a membrane, the axolemma. A cytoskeleton consisting of neurofilaments and microtubules runs through the axon. The microtubules provide a framework for fast axonal transport (see Axonal Transport section). Specialized molecular motors (kinesin molecules) bind to vesicles containing molecules (eg, neurotransmitters) destined for transport via a series of adenosine triphosphate (ATP)-consuming steps along the microtubules.

The axon conducts electrical signals from the initial segment (the proximal part of the axon, near the cell body) to the synaptic terminals. The initial segment has distinctive morphological features; it differs from both cell body and axon. The axolemma of the initial segment contains a high density of sodium channels, which permit the initial segment to act as a trigger zone. In this zone, action potentials are generated so that they can travel along the axon, finally invading the terminal axonal branches and triggering synaptic activity, which impinges on other neurons. The initial segment does not contain Nissl substance (see Fig 2–3). In large neurons, the initial segment arises conspicuously from the axon hillock, a cone-shaped portion of the cell body. Axons range in length from a few microns (in interneurons) to well over a meter (ie, in a lumbar motor neuron that projects from the spinal cord to the muscles of the foot) and in diameter from 0.1 μm to more than 20 μm.

A. Myelin

Many axons are covered by myelin. The myelin consists of multiple concentric layers of lipid-rich membrane produced by Schwann cells in the peripheral nervous system (PNS) and by oligodendrocytes (a type of glial cell) in the central nervous system (CNS) (Figs 2–6, 2–7, 2–8, 2–9, 2–10). The myelin sheath is divided into segments about 1 mm long by small gaps (1 μm long) where myelin is absent; these are the nodes of Ranvier. The smallest axons are unmyelinated. As noted in Chapter 3, myelin functions as an insulator. In general, myelination serves to increase the speed of impulse conduction along the axon.

B. Axonal Transport

In addition to conducting action potentials, axons transport materials from the cell body to the synaptic terminals (anterograde transport) and from the synaptic terminals to the cell body (retrograde transport). Because ribosomes are not present in the axon, new protein must be synthesized and moved to the axon. This occurs via several types of axonal transport, which differ in terms of the rate and the material transported. Anterograde transport may be fast (up to 400 mm/d) or slow (about 1 mm/d). Retrograde transport is similar to rapid anterograde transport. Fast transport involves microtubules extending through the cytoplasm of the neuron.

An axon can be injured by being cut or severed, crushed, or compressed. After injury to the axon, the neuronal cell body responds by entering a phase called the axon reaction, or chromatolysis. In general, axons within peripheral nerves can regenerate quickly after they are severed, whereas those within the CNS do not tend to regenerate. The axon reaction and axonal regeneration are further discussed in Chapter 22.

Synapses

Transmission of information between neurons occurs at synapses. Communication between neurons usually occurs from the axon terminal of the transmitting neuron (presynaptic side) to the receptive region of the receiving neuron (postsynaptic side) (Figs 2–6 and 2–11). This specialized interneuronal complex is a synapse, or synaptic junction. As outlined in Table 2–1, some synapses are located between an axon and a dendrite (axodendritic synapses, which tend to be excitatory), or a thorn, or mushroom-shaped dendritic spine which protrudes from the dendrite (Fig 2–12). Other synapses are located between an axon and a nerve cell body (axosomatic synapses, which tend to be inhibitory). Still other synapses are located between an axon terminal and another axon; these axoaxonic synapses modulate transmitter release by the postsynaptic axon. Synaptic transmission permits information from many presynaptic neurons to converge on a single postsynaptic neuron. Some large cell bodies receive several thousand synapses (see Fig 2–4).

Impulse transmission at most synaptic sites involves the release of a chemical transmitter substance (see Chapter 3); at other sites, current passes directly from cell to cell through specialized junctions called electrical synapses, or gap junctions. Electrical synapses are most common in invertebrate nervous systems, although they are found in a small number of sites in the mammalian CNS. Chemical synapses have several distinctive characteristics: synaptic vesicles on the presynaptic side, a synaptic cleft, and a dense thickening of the cell membrane on both the receiving cell and the presynaptic side (see Fig 2–11). Synaptic vesicles contain neurotransmitters, and each vesicle contains a small packet, or quanta, of transmitter. When the synaptic terminal is depolarized (by an action potential in its parent axon), there is an influx of calcium. This calcium influx leads to phosphorylation of a class of proteins called synapsins. After phosphorylation of synapsins, synaptic vesicles dock at the presynaptic membrane facing the synaptic cleft, fuse with it, and release their transmitter (see Chapter 3).

Synapses are very diverse in their shapes and other properties. Some are inhibitory and some excitatory; in some, the transmitter is acetylcholine; in others, it is a catecholamine, amino acid, or other substance (see Chapter 3). Some synaptic vesicles are large, some small; some have a dense core, whereas others do not. Flat synaptic vesicles appear to contain an inhibitory mediator; dense-core vesicles contain catecholamines.

In addition to calcium-dependent, vesicular neurotransmitter release, there is also a second, nonvesicular mode of neurotransmitter release that is not calcium-dependent. This mode of release depends on transporter molecules, which usually serve to take up transmitter from the synaptic cleft.

Nerve cell bodies are grouped characteristically in many parts of the nervous system. In the cerebral and cerebellar cortices, cell bodies aggregate to form layers called laminas. Nerve cell bodies in the spinal cord, brain stem, and cerebrum form compact groups, or nuclei. Each nucleus contains projection neurons, whose axons carry impulses to other parts of the nervous system, and interneurons, which act as short relays within the nucleus. In the peripheral nervous system, these compact groups of nerve cell bodies are called ganglia.

Groups of nerve cells are connected by pathways formed by bundles of axons. In some pathways, the axon bundles are sufficiently defined to be identified as tracts, or fasciculi; in others, there are no discrete bundles of axons. Aggregates of tracts in the spinal cord are referred to as columns, or funiculi (see Chapter 5). Within the brain, certain tracts are referred to as lemnisci. In some regions of the brain, axons are intermingled with dendrites and do not run in bundles so that pathways are difficult to identify. These networks are called the neuropil (Fig 2–13).

Neuroglial cells, commonly called glial cells, outnumber neurons in the brain and spinal cord 10:1. They do not form synapses. These cells appear to play a number of important roles, including myelin formation, guidance of developing neurons, maintenance of extracellular K⁺ levels, and reuptake of transmitters after synaptic activity. There are two broad classes of glial cells, macroglia and microglia (Table 2–2).

Macroglia

The term macroglia refers to astrocytes and oligodendrocytes, both of which are derived from ectoderm. In contrast to neurons, these cells may have the capability, under some circumstances, to regenerate.

Astrocytes

There are two broad classes of astrocytes: protoplasmic and fibrous. Protoplasmic astrocytes are more delicate, and their many processes are branched. They occur in gray matter. Fibrous astrocytes are more fibrous, and their processes (containing glial fibrils) are seldom branched. Astrocytic processes radiate in all directions from a small cell body. They surround blood vessels in the nervous system, and they cover the exterior surface of the brain and spinal cord below the pia.

Astrocytes provide structural support to nervous tissue and act during development as guidewires that direct neuronal migration. They also maintain appropriate concentrations of ions such as K⁺ within the extracellular space of the brain and spinal cord. Astrocytes may also play a role in synaptic transmission. Many synapses are closely invested by astrocytic processes, which appear to participate in the reuptake of neurotransmitters. Astrocytes also surround endothelial cells within the CNS, which are joined by tight junctions that impede the transport of molecules across the capillary epithelium, and contribute to the formation of the blood–brain barrier (see Chapter 11). Although astrocytic processes around capillaries do not form a functional barrier, they can selectively take up materials to provide an environment optimal for neuronal function.

Astrocytes form a covering on the entire CNS surface and proliferate to aid in repairing damaged neural tissue (Fig 2–14). These reactive astrocytes are larger, are more easily stained, and can be definitively identified in histological sections because they contain a characteristic, astrocyte-specific protein: glial fibrillary acidic protein (GFAP). Chronic astrocytic proliferation leads to gliosis, sometimes called glial scarring. Whether glial scarring is beneficial, or inhibits regeneration of injured neurons, is currently being studied.

Oligodendrocytes

Oligodendrocytes predominate in white matter; they extend arm-like processes which wrap tightly around axons, extruding the oligodendroglial cytoplasm to form a compact sheath of myelin which acts as an insulator around axons in the CNS. Oligodendrocytes may also provide some nutritive support to the neurons they envelop. A single oligodendrocyte may wrap myelin sheaths around many (up to 30–40) axons (see Figs 2–9 and 2–10). In peripheral nerves, by contrast, myelin is formed by Schwann cells. Each Schwann cell myelinates a single axon, and remyelination can occur at a brisk pace after injury to the myelin in the peripheral nerves.

Microglia

Microglial cells are the macrophages, or scavengers, of the CNS. They constantly survey the brain and spinal cord, acting as sentries to detect, and destroy, invaders (such as bacteria). When an area of the brain or spinal cord is damaged or infected, microglia activate and migrate to the site of injury to remove cellular debris. Some microglia are always present in the brain, but when injury or infection occurs, others enter the brain from blood vessels. Microglia play an important role in protecting the nervous system from outside invaders such as bacteria. Their role after endogenous insults, including stroke or neurodegenerative diseases such as Alzheimer disease, is under investigation.

Extracellular Space

There is some fluid-filled space between the various cellular components of the CNS. This extracellular compartment probably accounts for, under most circumstances, about 20% of the total volume of the brain and spinal cord. Because transmembrane gradients of ions, such as K⁺ and Na⁺, are important in electrical signaling in the nervous system (see Chapter 3), regulation of the levels of these ions in the extracellular compartment (ionic homeostasis) is an important function, which is, at least in part, performed by astrocytes. The capillaries within the CNS are completely invested by glial or neural processes. Moreover, capillary endothelial cells in the brain (in contrast to capillary endothelial cells in other organs) form tight junctions, which are impermeable to diffusion, thus creating a blood–brain barrier. This barrier isolates the brain extracellular space from the intravascular compartment.

Clinical Correlation

In cerebral edema, there is an increase in the bulk of the brain. Cerebral edema can be either vasogenic (primarily extracellular) or cytotoxic (primarily intracellular). Because of the limited size of the cranial vault within the skull, cerebral edema must be treated emergently.

The neuronal cell body maintains the functional and anatomic integrity of the axon (Fig 2–14). If the axon is cut, the part distal to the cut degenerates (wallerian degeneration), because materials for maintaining the axon (mostly proteins) are formed in the cell body and can no longer be transported down the axon (axoplasmic transport).

Distal to the level of axonal transection when a peripheral nerve is injured, Schwann cells dedifferentiate and divide. Together with macrophages, they phagocytize the remnants of the myelin sheaths, which lose their integrity as the axon degenerates.

After injury to its axon, the neuronal cell body exhibits a distinct set of histological changes (which have been termed the axon reaction or chromatolysis). The changes include swelling of the cell body and nucleus, which is usually displaced from the center of the cell to an eccentric location. The regular arrays of ribosome-studded endoplasmic reticulum, which characterize most neurons, are dispersed and replaced by polyribosomes. (The ribosome-studded endoplasmic reticulum, which had been termed the Nissl substance by classical neuroanatomists, normally stains densely with basic dyes. The loss of staining of the Nissl substance, as a result of dispersion of the endoplasmic reticulum during the axon reaction, led early scientists to use the term "chromatolysis.") In association with the axon reaction in some CNS neurons, there is detachment of afferent synapses, swelling of nearby astrocytes, and activation of microglia. Successful axonal regeneration does not commonly occur after injury to the CNS. Many neurons appear to be dependent on connection with appropriate target cells; if the axon fails to regenerate and form a new synaptic connection with the correct postsynaptic cells, the axotomized neuron may die or atrophy.

Regeneration

A. Peripheral Nerves

Regeneration denotes a nerve's ability to regrow to an appropriate target, including the reestablishment of functionally useful connections (see Figs 2–14 and 2–15). Shortly (1–3 days) after an axon is cut, the tips of the proximal stumps form enlargements, or growth cones. The growth cones send out exploratory pseudopodia that are similar to the axonal growth cones formed in normal development. Each axonal growth cone is capable of forming many branches that continue to advance away from the site of the original cut. If these branches can cross the scar tissue and enter the distal nerve stump, successful regeneration with restoration of function may occur.

The importance of axonal regeneration through the Schwann cell tubes surrounded by basal lamina (Büngner's bands) in the distal stump explains the different degrees of regeneration that are seen after nerve crush compared with nerve transection. After a crush injury to a peripheral nerve, the axons may be severed, but the Schwann cells, surrounding basal lamina, and perineurium maintain continuity through the lesion, facilitating regeneration of axons through the injured nerve. In contrast, if the nerve is cut, the continuity of these pathways is disrupted. Even with meticulous surgery, it can be difficult to align the proximal and distal parts of each axon's pathway; successful regeneration is, therefore, less likely.

B. Central Nervous System

Axonal regeneration is typically abortive in the CNS. The reasons for regeneration failure are not yet entirely clear. Classical neuropathologists suggested that the glial scar, which is largely formed by astrocytic processes, may be partly responsible. The properties of the oligodendroglial cells (in contrast to those of the Schwann cells of peripheral nerves) may also account for the difference in regenerative capacity. An inhibitory factor produced by oligodendrocytes, CNS myelin, or both may interfere with regeneration of axons through the CNS. It is now appreciated that molecules such as NoGo act as "stop signs" that inhibit regeneration of axons within the brain and spinal cord. Neutralization of NoGo has been shown to promote the regeneration of axons within the spinal cord in experimental animals. When confronted with a permissive environment (eg, when the transected axons of CNS neurons are permitted to regrow into a peripheral nerve, or transplanted into the CNS as a "bridge"), CNS axons can regenerate for at least a few centimeters. Some of the regenerated axons can establish synaptic connections with appropriate target cells.

C. Remyelination

In a number of disorders of the peripheral nervous system (such as the Guillain–Barré syndrome), there is demyelination, which interferes with conduction (see Chapter 3). This condition is often followed by remyelination by Schwann cells, which are capable of elaborating new myelin sheaths in the peripheral nervous system. In contrast, remyelination occurs much more slowly (if at all) in the CNS. Little remyelination occurs within demyelinated plaques within the brain and spinal cord in multiple sclerosis. A different form of plasticity (ie, the molecular reorganization of the axon membrane that acquires sodium channels in demyelinated zones) appears to underlie clinical remissions (in which there is neurological improvement) in patients with multiple sclerosis.

D. Collateral Sprouting

This phenomenon has been demonstrated in the CNS as well as in the peripheral nervous system (see Fig 2–13). It occurs when an innervated structure has been partially denervated. The remaining axons then form new collaterals that reinnervate the denervated part of the end organ. This kind of regeneration demonstrates that there is considerable plasticity in the nervous system and that one axon can take over the synaptic sites formerly occupied by another.

It has classically been believed that neurogenesis—the capability for production of neurons from undifferentiated, proliferative progenitor cells—is confined to the development period that precedes birth in mammals. According to this traditional view, after pathological insults that result in neuronal death, the number of neurons is permanently reduced. However, recent evidence has indicated that a small number of neuronal precursor cells, capable of dividing and then differentiating into neurons, may exist in the forebrain of adult mammals, including humans. These rare precursor cells reside in the subventricular zone. For example, there is some evidence for postnatal neurogenesis in the dentate gyrus of the hippocampus, and it has been suggested that the rate of generation of new neurons in this critical region can be accelerated in an enriched environment. While the number of new neurons that can be produced within the adult human brain is still being debated, the existence of these precursor cells may suggest strategies for restoring function after injury to the CNS. This is an area of intense research.