Figure 32-1. A coronal section through the hemisphere (left) showing the major types of fibers projecting to and from the cerebral cortex. The representation on the right shows layers (I to VI) of the cerebral cortex as they appear after staining for cell bodies or for the myelin sheath.
Figure 32-2. Nissl (A) and myelin (B) stains of adjacent sections of the human cerebral cortex and a Golgi impregnation (C) of a pyramidal neuron in the primate neocortex. (A and B Courtesy of Drs. Grayzna Rajkowska and Patricia Goldman-Rakic, University of Mississippi Medical Center and Yale University. C Courtesy of Dr. José Rafols, Wayne State University.)
The pattern of distribution of neuron cell bodies is called cytoarchitecture. The cytoarchitecture of the cerebral cortex is characterized by a layered arrangement. Most of the cerebral cortex has six distinct layers of neurons and is classified as neocortex. Two regions of the cerebral cortex have fewer than six layers. The first contains only three layers, is classified as archicortex, and includes the hippocampal formation. The second contains three to five layers, is classified as paleocortex, and includes the olfactory sensory area and the nearby entorhinal and periamygdaloid cortices. The following discussion concentrates primarily on the neocortex.
The neuronal layers in the neocortex are designated by Roman numerals, beginning at the pial surface (Fig. 32-1). There are six layers in the neocortex, with some of these layers being further subdivided on the basis of their architectural features.
Layer I, the molecular layer, contains very few neuron cell bodies and consists primarily of axons running parallel (horizontal) to the surface of the cortex. The apical dendrites of cells located in deeper layers also ramify within layer I.
Layer II, the external granular layer, is composed of a mixture of small neurons called granule cells and slightly larger neurons that are called pyramidal cells on the basis of the shape of their cell body. The apical dendrites of these pyramidal cells extend into layer I and their axons descend into and through the deeper cortical layers.
Layer III, the external pyramidal layer, contains primarily small to medium-sized pyramidal cells along with some neurons of other types. In general, the smaller pyramidal cells are sequestered in the outer or superficial portion of layer III; the larger pyramidal cells are located in the inner or deeper portion of this layer. Their apical dendrites ascend into layer I and their axons descend into and through the deeper layers.
Layer IV, the internal granular layer, consists almost exclusively of smooth (aspiny) stellate (star-like) neurons and spiny stellate neurons, both of which have sometimes been categorized as granule cells. This layer is free of pyramid-shaped cells. It can be divided into outer (IVa) and inner (IVb) portions in many neocortical areas and into three portions (IVa, IVb, IVc) in the primary visual cortex. Layer IV is the primary target for ascending sensory information from the thalamus.
Layer V, the internal pyramidal layer, consists predominantly of medium to large pyramidal cells. Apical dendrites of the medium pyramidal cells may extend upward one or two layers, whereas those of the large pyramidal cells extend outward to layer I. The large pyramidal cells of this layer are a major source of cortical efferent fibers, including axons to the basal nuclei, brainstem, and spinal cord. Some corticocortical axons also originate in layer V. These are probably collateral branches of axons that are projecting to some subcortical target.
Layer VI, the multiform layer, contains an assortment of neuron types, including some with pyramidal and fusiform cell bodies. The dendrites of the larger cells extend into layer I; those arising from the smaller cells usually extend no farther than layer IV. The axons of the cells of this layer project to subcortical targets, such as the thalamus, and to other cortical regions as corticocortical connections.
Two features of the myelinated fibers in the neocortex are noteworthy. First, there are prominent plexuses of horizontally running myelinated fibers in layers IV and V. These are called the outer and inner bands of Baillarger, respectively (Fig. 32-1). In the primary visual cortex, bordering on the calcarine sulcus, the outer band of Baillarger is greatly expanded. This band can be seen with the naked eye in fresh and stained sections and is called the stria (line) of Gennari (see Fig. 20-20). Second, in most regions of the neocortex, there are many radially oriented bundles of axons passing between the subcortical white matter and various parts of the cortex or between inner and outer cortical layers (Fig. 32-2B).
A variety of neuroactive substances are associated with neurons of the cerebral cortex. Principal among these are glutamate, aspartate, and γ-aminobutyric acid (GABA). Pyramidal cells are the efferent neurons of the cerebral cortex. They are predominantly glutaminergic and are excitatory to their targets. Most interneurons within the cortex are GABAergic and are inhibitory. The pyramidal cells of the cortex and therefore the output of the cortex are modulated by a variety of cortical afferents. The influence of these afferent fibers is to act on pyramidal cells either directly or via interneurons. A variety of neuropeptides (monoamines) are also found in the cerebral cortex; they influence not only populations of neurons but also local metabolic activity and vascular smooth muscle. The most important monoamines in the cortex are (1) norepinephrine, which originates from the locus ceruleus of the pons and distributes sparsely to all cortical layers; (2) dopamine, which arises from the substantia nigra–pars compacta and the adjacent ventral tegmental area and is found in moderate amounts in layers I and VI and sparsely in layers II to V; and (3) serotonin, which arises from the raphe nuclei and distributes heavily to all cortical layers.
The most common type of neuron in the cerebral cortex is the pyramidal cell (Figs. 32-2 and 32-3A). Pyramidal cells are found in all layers of the cortex with the exception of the molecular layer (layer I), and they are the predominant cell type in layers II, III, and V (Fig. 32-4). Pyramidal cells are characterized by (1) a roughly triangular cell body; (2) a single large apical dendrite that arises from the apex of the cell body and usually extends toward the molecular layer, giving off branches along the way; (3) an array of basal dendrites that run in a predominantly horizontal direction; and (4) an axon that originates from the base of the soma, leaves the cortex, and passes through the white matter.
Figure 32-4. Representative cell types in the cerebral cortex and the layers in which their cell bodies and dendrites are found. Dendrites of pyramidal cells (Py) of layers II, III, and V extend into layer I, whereas those of modified pyramidal cells (mPy) in layer VI extend only to about layer IV. Chandelier cells (Ch) are restricted almost entirely to layer III. The somata of aspiny and spiny stellate neurons (Asp, Sp) are in layer IV, although their processes extend into other layers. Basket cells (Bas) have processes that collectively extend into all cortical layers from cell bodies located mainly in layers III and V. (Modified from Hendry SHC, Jones EG: Sizes and distributions of intrinsic neurons incorporating tritiated GABA in monkey sensory-motor cortex. J Neurosci 1:390-408, 1981; and Jones EG: Laminar distribution of cortical efferent cells. In Peters A, Jones EG [eds]: Cerebral Cortex, vol 1. New York, Plenum Press, 1984, pp 521-553, with permission.)
The cell bodies of most pyramidal neurons range in size from 10 to 50 µm in height. The largest, called giant pyramidal cells of Betz or Betz cells, are found almost exclusively in the primary motor cortex, which is located in the precentral and anterior paracentral gyri. Their somata may reach 100 µm in height. Betz cells are most common in the region of motor cortex that projects to the anterior horn of the lumbar spinal cord and hence are concerned with the control of leg movement. These cells are so large that they can be distinguished with the naked eye in Nissl-stained sections of the human brain.
Both apical and basal dendrites of pyramidal cells are characterized by membrane specializations called dendritic spines. These spines are small outgrowths from the dendrite that give the impression of thorns on a rosebush (Fig. 32-3). The vast majority of synaptic contacts received by a pyramidal cell are located on dendritic spines rather than directly on the dendrite shaft or on the cell body.
Pyramidal neurons represent virtually the only output pathway for the cerebral cortex. Almost all other cell types in the cortex are local circuit neurons that exert their influence within their own immediate vicinity. Axons of pyramidal cells may terminate in another region of the cortex in the same hemisphere (association fibers), decussate in the corpus callosum to terminate in the cerebral cortex of the opposite hemisphere (callosal fibers), or course through the white matter to any of the numerous subcortical targets in the forebrain, brainstem, or spinal cord (projection fibers).
Pyramidal cells display a laminar organization, with the cell bodies in a given layer projecting to specific neural targets (Fig. 32-4). In general, pyramidal neurons in layers II and III give rise to association and callosal fibers. Pyramidal cells in layer V project to many subcortical structures, including the spinal cord, as projection fibers. The neurons in layer VI send their axons to a variety of locations, including thalamic nuclei and other regions of cortex. Within the cortex, axons of pyramidal cells send off an extensive and relatively dense array of axon collaterals. These collaterals terminate in all cortical layers and extend through a horizontal area covering several millimeters around the cell body (Fig. 32-5).
Figure 32-5. The cell bodies and dendrites (in red) of three pyramidal cells in the cerebral cortex compared with the intracortical distribution of axons (in blue) arising from these cells. Axon collaterals distribute over a much wider area than do the dendrites arising from the same cell. (Modified from Scheibel ME, Scheibel AB: Elementary processes in selected thalamic and cortical subsystems—the structural substrates. In Schmitt FO: The Neurosciences: Second Study Program, vol 2. New York, Rockefeller University Press, 1970, pp 443-457, with permission.)
As mentioned previously, all the various nonpyramidal neurons of the cerebral cortex function as cortical interneurons; that is, their axons do not leave the immediate region of the cell body. These cells are often referred to as local circuit neurons or intrinsic cortical neurons.
Santiago Ramón y Cajal, working in the late 1800s and early 1900s, described a rich variety of intrinsic cortical neurons. However, by the 1950s, it had become customary to refer to virtually all intrinsic cortical neurons as stellate cells, even though many were not actually star shaped. Now the pendulum has swung in the other direction, and a number of distinct morphologic types are recognized. Some of the more important of these are illustrated in Figure 32-4: spiny and aspiny stellate cells, basket cells, and chandelier cells.
Three types of intrinsic neurons receive thalamocortical axon terminals in layer IV: the small spiny cells, the aspiny stellate cells, and dendrites of the large basket cells. Of these, the spiny cells are believed to be excitatory, whereas basket cells and aspiny stellate cells use the neurotransmitter GABA and are thus considered to be inhibitory interneurons. Most other intrinsic neurons are presumed to be inhibitory. On the other hand, pyramidal neurons are uniformly associated with excitatory neurotransmitters, glutamate and aspartate in particular.
The basic framework of the internal circuit diagram of small regions of the cerebral cortex is well understood. In contrast, the details of this circuitry are only partially known and are in fact so complex as to defy the construction of a detailed circuit diagram like that used to represent a computer’s electronic hardware. For example, a single axon may branch repeatedly and contact hundreds of other neurons. A single neuron may also receive synaptic contacts from thousands of other neurons. Within a small volume of cortex, there may be millions of neurons.
The basic framework of cortical circuitry consists of afferent fibers, local circuits for the processing of this afferent information, and efferent fibers that convey the processed information to another site (Fig. 32-6). Thalamocortical axons terminate primarily in layer IV and to a lesser extent in layers III and VI. In layer IV, they terminate on excitatory and inhibitory interneurons as well as on dendrites from neurons in other layers (Fig. 32-4). The axons of interneurons in turn may end on dendrites of pyramidal cells or of other interneurons. The local processing of information culminates in connections to pyramidal cells, which carry the information to other cortical or subcortical regions. A copy of the information also goes to neurons in the immediate vicinity via axon collaterals (Fig. 32-5).
Figure 32-6. Basic circuits in the cerebral cortex. Afferent fibers are shown in blue and gray, interneurons in green, and efferent fibers in red. Thalamocortical fibers terminate primarily in layer IV, whereas corticocortical fibers and diffuse cortical afferents synapse in all layers. Pyramidal cells in the outer layers give rise to corticocortical projections, and those in layer V project to a wide range of subcortical targets.
The general pattern of termination of corticocortical axons is quite different from that of thalamocortical axons. Corticocortical axons branch repeatedly and make synaptic contacts on neurons in all layers of the cortex (Fig. 32-4).
The cerebral cortex receives a third set of inputs, called diffuse inputs, which consists of fibers that branch extensively and end diffusely over a wide area of cortex without respect for cytoarchitectural boundaries (Fig. 32-6). These inputs arise from a variety of sources, including certain nonspecific nuclei of the thalamus (e.g., the ventral anterior, central lateral, and midline nuclei), the locus ceruleus, and the basal nucleus (of Meynert). These structures are generally concerned with regulation of overall levels of cortical excitability and the associated phenomena of arousal, sleep, and wakefulness.
The cytoarchitecture of the cortex differs from one area to another in ways that are related to function (Fig. 32-7). In primary sensory cortex, layer IV, the major input layer of cortex, is especially thick, whereas layer V, the major projection layer, is narrow and indistinct. Cortex with this pattern is called heterotypical granular cortex. In primary motor cortex, the pattern is reversed: layer IV is almost invisible, and layer V is very thick, seeming to merge directly with layer III. Thus the projection layer is prominent and the input layer is small. Cortex of this type is called heterotypical agranular cortex. In most other areas of the neocortex, including the association cortices, the six layers are all clearly represented and are of roughly equal thickness. This type of cortex is called homotypical.
Figure 32-7. Typical cytoarchitectural patterns for homotypical, heterotypical granular, and heterotypical agranular regions. (Modified from Campbell AW: Histological Studies on the Localization of Cerebral Function. Cambridge, Cambridge University Press, 1905.)
The cerebral cortex has been subdivided on the basis of cytoarchitectural differences by many different investigators. The most famous of these, Korbinian Brodmann, worked in the early part of the twentieth century. He identified 47 distinct areas (Fig. 32-8), and his numbering scheme is still in common use today in both research and clinical settings. For example, the primary visual cortex is Brodmann area 17, and the primary motor cortex is area 4. In most instances, Brodmann cytoarchitectural areas are coextensive with cortical regions that have specific functional characteristics.
Figure 32-8. Cytoarchitectural map showing Brodmann areas on the lateral (A) and medial (B) surfaces of the hemisphere. (After Brodmann K. Modified from Carpenter MB, Sutin J: Human Neuroanatomy. Baltimore, Williams & Wilkins, 1983, with permission.)
A second, vertical pattern of organization is superimposed on the horizontal layered pattern described earlier. Unlike the cortical layers, this vertical pattern is not immediately obvious in histologic sections stained for neuron cell bodies (Nissl stains). However, when Golgi-stained material is studied, it is clear that neurons are often grouped together so that their cell bodies, axons, and apical dendrites form clusters that are oriented at right angles to the surface of the cortex.