Medical Physiology, 3rd Edition

Cells of the Nervous System

The neuron doctrine first asserted that the nervous system is composed of many individual signaling units—the neurons

In 1838, Schleiden and Schwann proposed that the nucleated cell is the fundamental unit of structure and function in both plants and animals. They reached this conclusion by microscopic observation of plant and animal tissues that had been stained to reveal their cellular composition. However, the brain proved to be more difficult to stain than other tissues, and until 1885, when Camillo Golgi introduced his silver-impregnation method, “the black reaction,” there was no clear indication that the brain is composed of individual cells. The histologist Santiago Ramón y Cajal worked relentlessly with the silver-staining method and eventually concluded that not only is nervous tissue composed of individual cells, but the anatomy of these cells also confers a functional polarization to the passage of nervous signals; the tapering branches near the cell body are the receptive end of the cell, and the long-axis cylinder conveys signals away from the cell. In the absence of any reliable physiological evidence, Ramón y Cajal was nevertheless able to correctly anticipate how complex cell aggregates in the brain communicate with each other.

The pathologist Heinrich von Waldeyer referred to the individual cells in the brain as neurons. He wrote a monograph in 1891 that assembled the evidence in favor of the cellular composition of nervous tissue, a theory that became known as the neuron doctrine. It is ironic that Golgi, whose staining technique made these advances possible, never accepted the neuron doctrine, and he argued vehemently against it when he received his Nobel Prize along with Ramón y Cajal in 1906. image N10-1 The ultimate proof of the neuron doctrine was established by electron microscopic observations that definitively demonstrated that neurons are entirely separate from one another, even though their processes come into very close contact.


Santiago Ramón y Cajal

For more information about Santiago Ramón y Cajal and the work that led to his Nobel Prize, visit (accessed October 2014).

Nerve cells have four specialized regions: cell body, dendrites, axon, and presynaptic terminals

Neurons are specialized for sending and receiving signals, a purpose reflected in their unique shapes and physiological adaptations. The structure of a typical neuron can generally be divided into four distinct domains: (1) the cell body, also called the soma or perikaryon; (2) the dendrites; (3) the axon; and (4) the presynaptic terminals (Fig. 10-1). The shape and organelle composition of these domains depends strongly on their cytoskeleton, which consists of three fibrillary structures: neurofilaments (i.e., intermediate filaments; see p. 23), microtubules (see pp. 23–25), and thin filaments (see pp. 25–28). The cytoskeleton—especially the microtubules and thin filaments, is dynamic and imbues axons and dendrites with the capacity to change shape, a plasticity believed to participate in the synaptic alterations associated with learning and memory.


FIGURE 10-1 Morphology of a typical neuron.

Cell Body

As the name perikaryon implies, the cell body is the portion of the cell surrounding the nucleus. It contains much of the cell's complement of endoplasmic reticular membranes as well as the Golgi complex. The cell body appears to be responsible for many of the neuronal housekeeping functions, including the synthesis and processing of proteins.


These structures are tapering processes of variable complexity that arise from the cell body. Dendrites, and to a lesser extent the cell body, are the main areas for receiving information. Thus, their membranes are endowed with receptors that bind and respond to the neurotransmitters released by neighboring cells. The chemical message is translated by membrane receptors into an electrical or a biochemical event that influences the state of excitability or function of the receiving neuron. The cytoplasm of the dendrites contains dense networks of microtubules as well as extensions of the endoplasmic reticulum.


Perhaps the most remarkable feature of the neuron, the axon is a projection that arises from the cell body, like the dendrites. Its point of origin is a tapered region known as the axon hillock. Just distal to the cone-shaped hillock is an untapered, unmyelinated region known as the initial segment. This area is also called the spike initiation zone because it is where an action potential (see p. 176) normally arises as the result of the electrical events that have occurred in the cell body and dendrites. In contrast to the dendrites, the axon is thin, does not taper, and can extend for more than a meter. Because of its length, the typical axon contains much more cytoplasm than does the cell body, up to 1000 times as much. The neuron uses special metabolic mechanisms to sustain this unique structural component. The cytoplasm of the axon, the axoplasm, is packed with parallel arrays of microtubules and neurofilaments that provide structural stability and a means to rapidly convey materials back and forth between the cell body and the axon terminus. Axons are self-reliant in energy metabolism, taking up glucose and oxygen from their immediate environment to produce ATP. Specialized glial cells called oligodendrocytes contribute in complex ways to axon integrity (see pp. 292–293).

Axons are the message-sending portion of the neuron. The axon carries the neuron's signal, the action potential, to a specific target, such as another neuron or a muscle. Some axons have a special electrical insulation, called myelin, that consists of the coiled cell membranes of glial cells that wrap themselves around the nerve axon (see pp. 292–293). If the axon is not covered with myelin, the action potential travels down the axon by continuous propagation. On the other hand, if the axon is myelinated, the action potential jumps from one node of Ranvier (the space between adjacent myelin segments) to another in a process called saltatory conduction (see pp. 200–201). This adaptation greatly speeds impulse conduction.

Presynaptic Terminals

At its target, the axon terminates in multiple endings—the presynaptic terminals—usually designed for rapid conversion of the neuron's electrical signal into a chemical signal. When the action potential reaches the presynaptic terminal, it causes the release of chemical signaling molecules in a complex process called synaptic transmission (see Chapters 8 and 13).

The junction formed between the presynaptic terminal and its target is called a chemical synapse. Synapse is derived from the Greek for “joining together” or “junction”; this word and concept were introduced in 1897 by the neurophysiologist Charles Sherrington, whose contributions led to a share of the 1932 Nobel Prize in Medicine or Physiology. image N10-2 A synapse comprises the presynaptic terminal, the membrane of the target cell (postsynaptic membrane), and the space between the two (synaptic cleft). In synapses between two neurons, the presynaptic terminals primarily contact dendrites and the cell body. The area of the postsynaptic membrane is frequently amplified to increase the surface that is available for receptors. This amplification can occur either through infolding of the plasma membrane or through the formation of small projections called dendritic spines.


Sir Charles Scott Sherrington

For more information about Sir Charles Scott Sherrington and the work that led to his Nobel Prize, visit (accessed October 2014).

The molecules released by the presynaptic terminals diffuse across the synaptic cleft and bind to receptors on the postsynaptic membrane. The receptors then convert the chemical signal of the transmitter molecules—either directly or indirectly—back into an electrical signal.

In many ways, neurons can be thought of as highly specialized endocrine cells. They package and store hormones and hormone-like molecules, which they release rapidly into the extracellular space by exocytosis (see pp. 34–35) in response to an external stimulus, in this case a nerve action potential. However, instead of entering the bloodstream to exert systemic effects, the substances secreted by neurons act over the very short distance of a synapse to communicate locally with a single neighboring cell (see p. 47).

In a different sense, neurons can be thought of as polarized cells with some of the properties of epithelial cells. Like epithelial cells, neurons have different populations of membrane proteins at each of the distinct domains of the neuronal plasma membrane, an arrangement that reflects the individual physiological responsibilities of these domains. Thus, the design of the nervous system permits information transfer across synapses in a selective and coordinated way that serves the needs of the organism and summates to produce complex behavior.

Synapses can undergo long-term changes based on certain patterns of prior activity. This plasticity (see pp. 328–333) is believed to underlie learning and memory.

The cytoskeleton helps compartmentalize the neuron and also provides the tracks along which material travels between different parts of the neuron

Neurons are compartmentalized in both structure and function. Dendrites are tapered, have limited length, and contain neurotransmitter receptor proteins in their membranes. Axons can be very long and have a high density of Na+ channels. Dendrites and the cell body contain messenger RNA, ribosomes, and a Golgi apparatus.

How does this compartmentalization come about? The molecular orientation of microtubules (see pp. 23–25) together with microtubule-associated proteins (MAPs) play important roles in dictating the vectorial transport of organelles and proteins. (Note that these MAPs are totally unrelated to the mitogen-activated protein [MAP] kinase; see p. 69.) Two major classes of MAPs are found in the brain: high-molecular-weight proteins such as MAP-1 and MAP-2 and lower-molecular-weight tau proteins. Both classes of MAPs associate with microtubules and help link them to other cell components. MAP-2 is most abundant in dendrites, where it may assist in dendrite formation. Tau proteins are confined to axons; their suppression in cultured neurons prevents formation of the axon without altering formation of the dendrites. Hyperphosphorylated tau proteins can assemble into pathological aggregates called neurofibrillary tangles, which are a hallmark of Alzheimer Disease.

The polarization of microtubules (see pp. 23–25) helps to create remarkable morphological and functional divisions in neurons. In axons, microtubules assemble with their plus ends pointed away from the cell body; this orientation polarizes the flow of material into and out of the axon. The cytoskeletal “order” provided by the microtubules and the MAPs helps define what should or should not be in the axonal cytoplasm. In dendrites, microtubules do not have a consistent orientation, which gives dendrites a greater functional similarity to the soma and perhaps their tapered shape.

The neuron cell body is the main manufacturing site for the membrane proteins and membranous organelles that are necessary for the structural integrity and function of its processes. Axons have little or no intrinsic protein synthetic ability, whereas dendrites have some free ribosomes and are able to engage in limited protein production. The transport of proteins from the cell body all the way to the end of long axons is a challenging task. The neuron also has a second task: moving various material in the opposite direction, from presynaptic terminals at the end of the axon to the cell body. The neuron solves these problems by using two distinct mechanisms for moving material to the presynaptic terminals in an “anterograde” direction and a third mechanism for transport in the opposite or “retrograde” direction (Table 10-2).

TABLE 10-2

Features of Axoplasmic Transport


SPEED (mm/day)



Fast anterograde


Saltatory movement along microtubules via the motor molecule kinesin (ATP dependent)

Vesicles containing peptides and other neurotransmitters, some degradative enzymes

Fast retrograde


Saltatory movement along microtubules via the motor molecule MAP-1C (brain dynein, ATP dependent)

Degraded vesicular membrane
Absorbed exogenous material (toxins, viruses, growth factors)

Slow (anterograde only)


Not clear; requires intact microtubules (ATP dependent)

Cytoskeletal elements (e.g., neurofilament and microtubule subunits)
Soluble proteins of intermediary metabolism

Fast Axoplasmic Transport

If the flow of materials from the soma to the distant axon terminus were left to the whims of simple diffusion, their delivery would be far too slow to be of practical use. It could take months for needed proteins to diffuse to the end of an axon, and the presynaptic terminals are high-volume consumers of these molecules. To overcome this difficulty, neurons exploit a rapid, pony express–style system of conveyance known as fast axoplasmic transport (see Table 10-2). Membranous organelles, including vesicles and mitochondria, are the principal freight of fast axoplasmic transport. The proteins (some associated with RNA), lipids, and polysaccharides that move at fast rates in axons do so because they have caught a ride with a membranous organelle (i.e., they are sequestered inside the organelle or bound to or inserted into the organellar membrane). The peptide and protein contents of dense-core secretory granules, which are found in the presynaptic axonal terminals, are synthesized as standard secretory proteins (see pp. 34–35). Thus, they are cotranslationally inserted across the membranes of the rough endoplasmic reticulum and subsequently processed in the cisternae of the Golgi complex. They are shipped to the axon in the lumens of Golgi-derived carrier vesicles.

Organelles and vesicles, and their macromolecule payloads, move along microtubules (see p. 23) with the help of a microtubule-dependent motor protein called kinesin (Fig. 10-2A). The kinesin motor (see p. 25) is itself an ATPase that produces vectorial movement of its payload along the microtubule. This system can move vesicles down the axon at rates of up to 400 mm/day; variations in cargo speed simply reflect more frequent pauses during the journey. Kinesins always move toward the plus end of microtubules (i.e., away from the cell body), and transport function is lost if the microtubules are disrupted. The nervous system has many forms of kinesin that recognize and transport different cargo. It is not known how the motor proteins recognize and attach to their intended payloads.


FIGURE 10-2 Fast axoplasmic transport. ER, endoplasmic reticulum.

Fast Retrograde Transport

Axons move material back toward the cell body via a different motor protein called brain dynein (see p. 25) or MAP-1C (see Fig. 10-2B). Like kinesin, MAP-1C moves along microtubule tracks and is an ATPase (see Table 10-2). However, MAP-1C moves along microtubules in the opposite direction to kinesin (see Fig. 10-2C), though at a slightly slower pace. Retrograde transport provides a mechanism for target-derived growth factors, like nerve growth factor, to reach the nucleus of a neuron where they can influence survival. How this signal is transmitted up the axon has been a persistent question. It may be endocytosed at the axon's terminal and transported to the cell body in a “signaling endosome.” The loss of ATP production in axons, as occurs with blockade of oxidative metabolism, causes fast axonal transport in both the anterograde and retrograde directions to fail.

Slow Axoplasmic Transport

Axons also have a need for hundreds of other proteins, including cytoskeletal proteins and soluble proteins that are used as enzymes for intermediary metabolism. These proteins are delivered by a slow anterograde axoplasmic transport mechanism that moves material at a mere 0.2 to 8 mm/day, the nervous system's equivalent of snail mail. The slowest-moving proteins are neurofilament and microtubule subunits (0.2 to 1 mm/day). The mechanism of slow axoplasmic transport is not well understood, but molecular motors operating on microtubule tracks appear to be involved. In fact, the difference between slow and fast axonal transport may primarily be the number of transport interruptions during the long axonal journey.

Neurons can be classified on the basis of their axonal projection, their dendritic geometry, and the number of processes emanating from the cell body

The trillions of nerve cells in the CNS have great structural diversity. Typically, neurons are classified on the basis of where their axons go (i.e., where they “project”), the geometry of their dendrites, and the number of processes that emanate from the cell body (Fig. 10-3). The real significance of these schemes is that they have functional implications.


FIGURE 10-3 Classification of neurons based on their structure.

Axonal Projection

Neurons with long axons that connect with other parts of the nervous system are called projection neurons (or principal neurons or Golgi type I cells). Each of these cells has a clearly defined axon that arises from the axon hillock located on the cell body or proximal dendrite and extends away from the cell body, sometimes for remarkable distances. Some neurons in the cortex, for example, project to the distal part of the spinal cord, a stretch of nearly a meter. All the other processes that a projection neuron has are dendrites. The other type of neuron that is defined in this way has all of its processes confined to one region of the brain. These neurons are called interneurons (or intrinsic neurons or Golgi type II cells). Some of these cells have very short axons, whereas others seem to lack a conventional axon altogether and may be referred to as anaxonal. The anaxonal neuron in the retina is called an amacrine cell (from the Greek for “no large/long fiber”).

Dendritic Geometry

A roughly pyramid-shaped set of dendritic branches characterizes pyramidal cells, whereas a radial pattern of dendritic branches defines stellate cells. This classification often includes mention of the presence or absence of dendritic spines—those small, protuberant projections that are sites for synaptic contact. All pyramidal cells appear to have spines, but stellate cells may have them (spiny) or may not (aspiny).

Number of Processes

Neurons can also be classified by the number of processes that extend from their cell bodies. The dorsal root ganglion cell is the classic unipolar neuron. The naming of the processes of primary sensory neurons, like the dorsal root ganglion cell, is often ambiguous. The process that extends into the CNS from this unipolar neuron is easily recognized as an axon because it carries information away from the cell body. On the other hand, the process that extends to sensory receptors in the skin and elsewhere is less easily defined. It is a typical axon in the sense that it can conduct an action potential, has myelin, and is characterized by an axonal cytoskeleton. However, it conveys information toward the cell body, which is usually the function of a dendrite. Bipolar neurons, such as the retinal bipolar cell, have two processes extending from opposite sides of the cell body. Most neurons in the brain are multipolar. Cells with many dendritic processes are designed to receive large numbers of synaptic inputs.

Most neurons in the brain can be categorized by two or more of these schemes. For example, the large neurons in the cortical area devoted to movement (i.e., the motor cortex) are multipolar, pyramidal, projection neurons. Similarly, a retinal bipolar cell is both an interneuron and a bipolar cell.

Glial cells provide a physiological environment for neurons

Glial cells are defined in part by what they lack: axons, action potentials, and synaptic potentials. They are more numerous than neurons and are diverse in structure and function. The main types of CNS glial cells are oligodendrocytes, astrocytes, and microglial cells. In the PNS, the main types of glial cells are satellite cells in autonomic and sensory ganglia, Schwann cells, and enteric glial cells. Glial function is discussed in Chapter 11. Oligodendrocytes form the myelin sheaths of CNS axons, and Schwann cells myelinate peripheral nerves. Glial cells are involved in nearly every function of the brain and are far more than simply “nerve glue,” a literal translation of the name neuroglia (from the Greek neuron [nerve] + glia [glue]).

In depictions of the nervous system, the presence of glial cells is sometimes minimized or neglected altogether. Glia fills in almost all the space around neurons, with a narrow extracellular space left between neurons and glial cells that has an average width of only ~0.02 µm. Glial cells have a major impact on the composition of the extracellular fluid, which in turn has a major impact on brain function, as we will see in Chapter 11.

Box 10-1

Neural Modalities

The type of information, or neural modality, that a neuron transmits is classically categorized by three terms that refer to different attributes of the neuron.

1. The first category defines the direction of information flow.

• Afferent (sensory): neurons that transmit information into the CNS from sensory cells or sensory receptors outside the nervous system. Examples are the dorsal root ganglion cell and neurons in the sensory nucleus of CN V.

• Efferent (motor): neurons that transmit information out of the CNS to muscles or secretory cells. Examples are spinal motor neurons and motor neurons in the ANS.

2. The second category defines the anatomical distribution of the information flow.

• Visceral: neurons that transmit information to or from internal organs or regions that arise embryologically from the branchial arch (e.g., chemoreceptors of the carotid body).

• Somatic: neurons that transmit information to or from all nonvisceral parts of the body, including skin and muscle.

3. The third category, which is somewhat arbitrary, defines the information flow on the basis of the embryological origin of the structure being innervated.

• Special: neurons that transmit information to or from a “special” subset of visceral or somatic structures. For example, in the case of special visceral neurons, information travels to or from structures derived from the branchial arch region of the embryo (e.g., pharyngeal muscles). In the case of special somatic neurons, which handle only sensory information, the neurons arise from the organs of special sense (e.g., retina, taste receptors, cochlea).

• General: neurons that transmit information to or from visceral or somatic structures that are not in the special group.

Each axon in the body conveys information of only a single modality. In this classification scheme, a motor neuron in the spinal cord is described as a general somatic efferent neuron. A motor neuron in the brainstem that innervates branchial arch–derived chewing muscles is described as a special visceral efferent neuron. Because each of these three categories defines two options, you might expect a total of eight distinct neural modalities. In practice, however, only seven neural modalities exist. The term special somatic efferent neuron is not used.