Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Cerebral hemispheres
The cerebral hemispheres are the largest part of the human brain: they consist of the telencephalon (cerebral cortex and underlying white matter), basal ganglia and diencephalon. Both hemispheres are interconnected by the corpus callosum and the anterior commissure. The cerebral cortex contains the motor and premotor regions, the highest level at which motor activities are controlled, and the primary sensory regions, which provide the first cortical level for the conscious experience of sensory stimuli and to which sensory organs project either directly (olfactory stimuli) or via specific thalamic nuclei. Cortical areas of higher levels are either modality-specific (higher unimodal cortices) or multimodal (association cortex), enabling complex analyses of the internal and external environments. The cortical part of the limbic system (hippocampal formation, amygdala, cingulate and retrosplenial cortex) is a multimodal region that is particularly concerned with memory and the emotional aspects of sensory experience and behaviour. It is also connected with subcortical areas such as the hypothalamus in integrating widespread physiological activities that include endocrine and vegetative reactions. Other association regions within the frontal, temporal and parietal lobes are concerned with the highest aspects of cognitive function, such as social communication, language and visuospatial attention. Anterior association regions of the frontal lobe contribute to personality, judgement, foresight and planning.
The white matter contains association fibres limited to each hemisphere; commissural fibres linking corresponding areas of both hemispheres; and projection fibres connecting the cerebral cortex with subcortical, brainstem and spinal cord nuclei. Some of these bundles (tracts, fasciculi) are relatively well defined and compact, while others fan out and are less easy to identify. A detailed knowledge of the three-dimensional anatomical interrelationships of white matter tracts is a requisite for the planning, intraoperative monitoring and execution of neurosurgical procedures for tumour surgery, resection of epileptic foci and deep brain stimulation. Current understanding of the fibre tracts owes much to the introduction of diffusion-weighted magnetic resonance imaging (dMRI), which enables both postmortem and in vivo visualization of the trajectory of the white matter tracts that connect distant regions (structural connectivity). Functional MRI (fMRI) defines the role of cortical areas by task-driven activation or resting state analyses of cortical networks, and thus allows the study of functional connectivity. Stereotaxic atlases in which data are registered to a reference brain are used to integrate different multimodal imaging findings and bridge between data acquired at different scales, from macroscopical in vivo MRI to microscopical postmortem anatomy ( , , , ).
Cerebral Hemisphere Surfaces, Sulci and GYRI
The external surface of each hemisphere is highly convoluted into a series of gyri, separated by sulci that represent extensions of the subarachnoid space ( Figs 32.1 – 32.2 32.3 ). The term fissure is used for the deepest sulci, whereas incisures or notches are short, shallow indentations. The main sulci have depths of 1–3 cm, and may harbour small annectant gyri connecting the walls of two adjacent gyri. Most sulci are orientated towards the nearest ventricular cavity and for this reason the sulcal floor is often used by neurosurgeons as point of entry into the underlying ventricle. The sulci of the adult brain are highly variable between subjects, and can be long or short, branching, continuous or interrupted by short transitional gyri ( , , , ). All long primary sulci can be consistently identified in most human brains (e.g. lateral fissure, central, precentral, postcentral, calcarine, parieto-occipital, collateral, cingulate and callosal sulci). Among the short sulci, primary sulci such as the rhinal, olfactory, lateral and occipital sulci are consistently found in all brains, whereas other short sulci are highly variable and composed of several branches (e.g. orbital and subparietal sulci), or are sometimes absent and therefore indicated as supplementary (e.g. middle frontal sulcus).
Each gyrus is composed of two lateral walls, and a crown that is visible on the external surface. It is estimated that about two-thirds of the cerebral cortex lies along the gyral walls and is therefore buried in the depths of the sulci ( , ). The majority of the sulci appear during postmenstrual weeks 18–21: cortical folding (gyrification) accelerates conspicuously from postmenstrual week 28 and reaches a transient maximum between the 7th and 13th months postnatally. From this time on, the degree of folding declines by 18% until the adult value will be reached at an age of approximately 23 years ( ).
The process of fetal cortical development includes sequential neural cell proliferation at the ventricular zone, ventriculofugal migration of neuronal progeny to form the most recent outer layer of the cortex, and increases in cortical thickness and expansion of the cortical surface area ( Ch. 14 ). The description of the external appearance of this process as a measurement of cerebral hemisphere growth and fetal age, i.e. the visualization of early surface ridges adjacent to neighbouring depressions, is fraught with limitations similar to those noted when describing the pre- and postnatal growth of individuals. Techniques to study hemispheric growth include examination of postmortem brains that have been fixed and removed from the cranium after spontaneous fetal death ( ); MR imaging of brains within the calvaria ( ); and ultrasound examination of routine second and third trimester scans of live fetuses ( , ). These studies have noted different ages at which sulci first appear, reflecting the heterogeneous ethnic, gender and nutritional status of the fetuses: in the case of postmortem studies, there is also the possibility that underlying pathology might have affected cerebral growth.
The main cerebral sulci divide the hemispheres into frontal, parietal, occipital, temporal, insular and limbic lobes ( Fig. 32.1A,B ). In the adult brain, the limbic lobe is visible on the medial surface of each hemisphere. Parts of the frontal, parietal and temporal lobes cover the insular lobe, which is present in the depth of the lateral fissure (opercularization). Each hemisphere has superolateral, medial and inferior (basal) surfaces and a frontal, temporal and occipital pole (see Fig. 32.1A,B ).
The superolateral surface is convex and lies beneath the bones of the calvaria: the frontal, parietal, temporal and occipital lobes correspond approximately in surface extent to the overlying bones from which they take their names. The frontal and parietal lobes are separated by the central (Rolandic) sulcus and both are separated from the temporal lobe by the prominent lateral (Sylvian) fissure. Frontal and temporal regions are each composed of three horizontal gyri, the superior, middle and inferior frontal and temporal gyri, respectively. The central region is composed of two slightly oblique gyri, the frontal precentral gyrus and the parietal postcentral gyrus, that are separated by the central sulcus. The deep intraparietal sulcus divides the superior parietal lobule from the inferior parietal lobule (which is subdivided into supramarginal and angular gyri). The occipital region is composed of three horizontal gyri (superior, middle and inferior occipital gyri), which are highly variable between subjects. The insula lies deep in the floor of the lateral fissure and consists of 4–5 diagonal gyri, separated by the central insular sulcus into anterior short and posterior long insular gyri.
The inferior surface has a smaller anterior frontal orbital part and a larger posterior tentorial part that belongs to the temporal and occipital lobes. The orbital part rests on the floor of the anterior cranial fossa, and the posterior part rests on the floor of the middle cranial fossa and the upper surface of the tentorium cerebelli. The orbital part of the inferior surface is made up of the orbital gyri and the gyrus rectus. The tentorial part of the inferior surface is covered by the basal aspects of the inferior temporal, inferior occipital and lingual gyri, and the fusiform gyrus.
The medial surface is flat and vertical, and is separated from the contralateral hemisphere by the longitudinal fissure and the falx cerebri. The medial surface of the hemisphere includes a C-shaped inner ring composed of two continuous gyri (the cingulate and parahippocampal gyri), surrounded by the gyrus rectus, superior frontal gyrus, paracentral lobule, precuneus, cuneus and lingual gyrus. Some details of the anatomy of the gyri and sulci of each lobe are illustrated in Figs 32.2 – 32.3 .
Neuronal Types in the Cerebral Cortex
The cerebral cortex is an intricate complex of neuronal somata and fibres, neuroglia and blood vessels. The neocortex contains two major neuronal cell types: excitatory pyramidal cells with long axons (projection neurones) and stellate cells ( Fig. 32.4 ). Pyramidal cells are the most abundant type and represent about 75% of the cortical neurones. Stellate cells have been subdivided into spiny stellate cells that express glutamate as neurotransmitter, and GABA-containing inhibitory aspinous interneurones with relatively short, locally terminating axons ( Fig. 32.5 ). All types have been further divided on the basis of their size and shape, connectivity, neurotransmitter and gene expression, and electrophysiological characteristics.
Pyramidal cells have a flask-shaped or triangular soma that ranges from 10 to 80 μm in diameter and gives rise to a single thick apical dendrite and multiple basal dendrites. The apical dendrite ascends towards the cortical surface, tapering and branching (see Fig. 32.4 ). Some dendrites end in the most superficial lamina, the molecular layer (layer I), others end at the level of the outer granular (layer II) or outer pyramidal layer (layer III). From the basal surface of the soma, dendrites spread horizontally or obliquely for distances up to 1 mm, depending on the size of the soma. Like apical dendrites, the basal dendrites branch profusely along their length. All pyramidal cell dendrites are studded with dendritic spines that become more numerous as the distance from the cell soma increases. A single slender axon arises from the axon hillock, which is usually situated centrally on the basal surface of a pyramidal neurone. For the vast majority, if not in all, pyramidal neurones, the axon leaves the cortical grey matter to enter the white matter. Before an axon reaches the white matter, it gives rise to a recurrent branch that terminates in the cortical region of origin of the parent axon. Pyramidal cells use an excitatory amino acid, either glutamate or aspartate, as neurotransmitter. Various types of pyramidal cells can be identified by their connectivity (e.g. axonal projections to the ipsi- or contralateral hemisphere, cortico-striate, cortico-thalamic, cortico-pontine and cortico-spinal projections), electrophysiological (e.g. regular fast or slow spiking, or intrinsically bursting types) and genetic characteristics. The giant Betz cells and the solitary Meynert cells are very large pyramidal cells with special morphological features.
The volume of the somata of Betz cells, located in lamina V of the primary motor cortex, is ten times larger than that of common layer V pyramidal cells. A Betz cell gives rise to a 10 μm thick myelinated axon that terminates on motor neurones in the brainstem and anterior horn of the spinal cord via the cortico-nuclear and cortico-spinal tracts, respectively. In contrast to common pyramidal cells, which have a single apical dendrite and dendritic arbors that exit the soma from basal angles, Betz cells have many dendrites that arise directly from the entire circumference of the soma. Meynert cells are located at the border between layers V and VI of the primary visual cortex; their somata are three times larger than those of common layer V pyramidal cells. In contrast to other pyramidal cells, most of their spines are found on the extensively arborized basal dendrites in layers V and VI, which extend below several blobs and interblobs (see below). Meynert cells are involved in motion detection, mediated via their connections with area MT/V5, and with the superior colliculus via axon collaterals. The von Economo neurone (spindle neurone) is a subtype of projection neurone that occurs preferentially in the deeper layer V of fronto-insular and anterior cingulate regions and has also been described in the entorhinal cortex, the hippocampal formation, and the dorsomedial Brodmann’s area 9 (BA9). These neurones are characterized by a single apical dendrite, a single basal dendrite and a long projection axon and are probably glutamatergic: their selective degeneration is found in early stages of fronto-temporal dementia.
Spiny stellate cells (see Fig. 32.4 ) are the second most numerous cell type in the neocortex. They are found preferentially in lamina IV and are the major target of thalamo-cortical input. Their relatively small multipolar somata commonly measure 6–10 μm in diameter. Several primary dendrites, profusely covered in spines, radiate for variable distances from each soma. The axons ramify within the grey matter, predominantly in the vertical plane. Spiny stellate cells use glutamate as neurotransmitter. These morphological and chemical characteristics led some authors to classify spiny stellates as a variant pyramidal cell, but their locally terminating axons also argues for a classification as interneurone.
All aspiny or sparsely spinous stellate cells are GABAergic (γ-aminobutyric acid) inhibitory interneurones (see Fig. 32.6 ). Their axons are confined to grey matter and terminate on other interneurones and pyramidal cells. Interneurones form inhibitory synapses with pyramidal neurones, and can also receive inhibitory synapses from other interneurones, but such connections are sparse: most synapses on interneurones are excitatory. There are many different subtypes of interneurones, including basket, horizontal, chandelier, double bouquet, bipolar/fusiform, bitufted, neurogliaform and Martinotti cells, and layer I interneurones (see Fig. 32.5 ; Fig. 32.6 ). Each morphological interneurone type can display widely varying electrophysiological or molecular properties. GABAergic interneurones often co-express neuropeptides (neuropeptide Y (NPY), somatostatin (SOM), cholecystokinin (CCK), vasoactive intestinal polypeptide (VIP)) and/or calcium binding proteins (parvalbumin (PV), calbindin (CB), calretinin (CR)) in a cell-type specific manner. GABAergic interneurones can be subdivided into several non-overlapping types according to differences in their developmental origin, target cells, and molecular markers. The embryonic medial and caudal ganglionic eminences are the origin of three GABAergic interneuronal types that co-express PV, SOM, and the serotonin receptor 5-HT 3 A . The PV type includes basket and chandelier cells: axons of PV basket cells terminate mainly in the perisomatic region, and chandelier cells co-express PV and CCK and terminate on the axon initial segment of pyramidal cells. The SOM type includes Martinotti cells that project to distal dendrites. The 5-HT 3 A type includes bipolar cells that co-express VIP and can lead to disinhibition (interneurone-selective disinhibitory cells, Fig. 32.7 ); Reelin cells (including Cajal–Retzius cells; see Fig. 32.5 and ); and CCK-expressing small basket cells. Neuronal nitric oxide synthase (nNOS) is constitutively expressed in 2% of cortical interneurones and is co-localized with SOM or NPY: these interneurones have high numbers of glutamate receptors that mediate the glutamate-induced hyperexcitation that follows cerebral impairment.
The common PV basket cells (see Fig. 32.6 ) are found in layers II–VI. They never express Reelin. Small basket cells express SOM, CCK and VIP. All basket cells show a mostly multipolar dendritic arbor, and a short, vertical axon that rapidly divides into horizontal collaterals that typically end in large terminal sprays on the somata and proximal dendrites of pyramidal cells, and rarely on other PV-expressing interneurones. The synaptic outputs of the fast-spiking (common) basket cells are strongly depressing.
The cell bodies of horizontal cells (see Fig. 32.6 ) lie mainly at the border between layers I and II, occasionally deep in lamina I (molecular layer). They are small and fusiform, and their dendrites spread short distances in layer I in two opposite directions. Their axons often stem from a dendrite, then divide into two branches that travel away from each other for great distances in the same layer. The morphological subtype of horizontal cells is heterogeneous with respect to the co-expression of PV, Reelin, SOM, VIP and NPY. Layer I horizontal cells express PV and Reelin, which is particularly important for the development of the regular lamination of the cortex during development. Other horizontal cells also express VIP, NPY and SOM.
Chandelier cells (see Fig. 32.5 ) are frequently found at the border between layers I/II and layer VI. Some, but not all, express PV. They have a variable morphology, although most somata are ovoid. Their dendrites arise from the upper and lower poles of the cell body and the axon emerges from either the cell body or a proximal dendrite. The ascending or descending axons ramify close to the cell body and terminate in numerous vertically orientated strings that run alongside the axon initial segment of pyramidal cells, where they build very effective inhibitory synapses.
Double bouquet cells (see Fig. 32.6 ) are found in laminae II and III and their axonal arborization is predominantly perpendicular to the pial surface, traversing layers II and V. Generally, these neurones have two or three main dendrites that terminate in a superficial and deep dendritic tuft. A single axon arises usually from the oval or spindle-shaped soma and rapidly divides into ascending and descending branches that collateralize extensively; the axonal arbor is confined to a perpendicularly extended, but horizontally confined, cylinder.
The bipolar cell (see Figs 32.5 – 32.6 ) has a very small, ovoid soma with a single ascending and a single descending dendrite that arise from the upper and lower poles, respectively of the soma. Sixty per cent of these cells are located in layers II and III. They co-express VIP, PV, CCK and acetycholine. They receive GABAergic input from basket cells on their somata, and from layer V pyramidal cells on their proximal dendrites. The dendrites branch sparsely and run vertically to produce a narrow dendritic tree that may extend through most of the cortical layers. Commonly, the axon originates from one of the primary dendrites and branches into a vertically elongated and horizontally confined axonal arbor that closely parallels the dendritic tree. Since pyramidal cells represent a major target for the GABA/PV/VIP synapses, the activation of these synapses may lead to coherent oscillations that provide the necessary signal to synchronize pyramidal cell discharges across and within cortical columns.
Bitufted cells (see Fig. 32.6 ) have two tufts of dendrites that originate directly from the soma and extend in vertical directions through all layers. The somata of these interneurones are most frequently located in layers II–IV and express Reelin. Some bitufted cells also express SOM. Layer II–III bitufted interneurones receive excitatory input from pyramidal cells in the same layers. Depression can be observed at these excitatory synapses: it may be caused by exocytotic GABA release from dendrites (since GABA-activated presynaptic GABA B receptors are found on the glutamatergic terminals), causing a reduction of glutamate release, i.e. conditioning. The axons of bitufted cells generate inhibitory synapses with pyramidal cells of the same cortical column (feedback inhibition). Since the elongated dendritic arbors span the entire cortex, bitufted interneurones receive input from all cortical layers, which means that bitufted cells can control the excitability of cortical columns.
The neurogliaform or spiderweb cell (see Fig. 32.5 ) is a small spherical cell found in layers I–IV. Most cells in layer I are neurogliaform cells that co-express PV, Reelin, NPY and VIP. Those in deeper layers also contain nNOS. Numerous short dendrites typically radiate out from their somata. The axon arises from the soma or a proximal dendrite and forms a wider, very dense axonal plexus composed of fine branches.
Most of the Martinotti cells (see Fig. 32.6 ) are located in supragranular layers II and III. They are more sparsely distributed throughout granular and infragranular layers. All express SOM and Reelin, but not all SOM-positive cells are Martinotti cells. They can also express CB, CR, CCK, VIP and NPY. The most striking feature of Martinotti cells in layers V–VI is their characteristic long ascending axon that spreads horizontally and arborizes extensively in layer I. Their bipolar or multipolar dendrites branch locally or down to deeper layers. Table 32.1 provides a summary of GABAergic interneurones, main targets and anatomical signatures.
TABLE 32.1
Isocortical GABAergic interneurones classified by their co-localization of parvalbumin, somatostatin and serotonin 5-HT 3 A receptors, morphological types, main targets and anatomical signatures
Adapted from .
| Main marker | Parvalbumin | Somatostatin | 5-HT 3 A | ||||||||
| % interneurones | 40% | 30% | 30% | ||||||||
| Morphological type | Basket cell | Chandelier cell | Martinotti cell | Non-Martinotti cell | VIP cell | Non-VIP cell | |||||
| 40% of 5-HT 3 A | 60% of 5-HT 3 A | ||||||||||
| Bipolar | CCK small basket | Multi polar | Small basket | CCK large basket | Neurogliaform cell | ||||||
| Main target | Soma and proximal dendrite | Only pyramidal cells; axon initial segment | Dendrites | Interneurones | Synaptic and volume transmission | ||||||
| Anatomical signature |
|
|
|
|
|
|
|
||||
These interneurones build the microcircuitry of the cerebral cortex with other interneurones and pyramidal (projection neurones) cells (see Fig. 32.7 ). They synaptically connect neurones of the same and/or different layers. Various sequences of direct or indirect synaptic connections between inhibitory interneurones and excitatory pyramidal cells are the basis of the complex modulatory mechanism of cortical microcircuitry. If GABAergic interneurones receive an excitatory input, they will inhibit the excitatory output of those cells by a direct synaptic connection with pyramidal cells (feedforward inhibition; Fig. 32.8 ). However, they will excite pyramidal cells if they are indirectly connected with them via an interposed inhibitory GABAergic cell that contacts the pyramidal cell (disinhibition). A third principal type of connectivity in microcircuits occurs if GABAergic interneurones receive an excitatory input from a pyramidal cell and send back an inhibitory output to the pyramidal cell (feedback inhibition). Impaired microcircuitry can lead to synchronized excitatory activity in cases of epilepsy.
Laminar organization
The most obvious microscopic feature of the cortex is its horizontal organization in layers or laminae that are visible to the naked eye or under the microscope after cell body or myelin staining. The isocortex has six layers in histological sections stained to show the somata (Nissl stain) ( Fig. 32.9 ). The laminar pattern of the allocortex is heterogeneous, showing less or more layers depending on the allocortical region. The most important features of myeloarchitectonic lamination are given in Table 32.2 .
TABLE 32.2
Most important criteria for myeloarchitectonic descriptions ( , )
From , modified
| Presence of Baillarger (layers 4 and 5b) stripes | ||
| Astriate | Baillarger stripes cannot be delineated due to the very high fibre density in layers 5a and 6a 1 Occurrence: large parts of the precentral gyrus |
|
| Propeastriate | A modification of the astriate type, because of a slight decrease of the fibre density in layers 5a and 6a, 1 i.e. the cortex is nearly astriate Occurrence: parts of the precentral gyrus, inferior parietal lobule, parts of the superior temporal gyrus |
|
| Propeunistriate | The outer Baillarger stripe is well demarcated, the inner stripe is less obvious, but better visible than in the unistriate type, since layer 6a is less dense than 5b. Fibre density in layer 5a is slightly lower than in the propeastriate type Occurrence: dorsal part of the medial frontal gyrus, parts of transverse and superior temporal gyri |
|
| Bistriate | Both Baillarger stripes clearly detectable because of lower fibre densities in layers 5b and 6a 1 Occurrence: frontal pole, medial frontal gyrus, parts of the transverse temporal gyrus, postcentral gyrus, superior parietal lobule, parts of extrastriate visual cortex |
|
| Unistriate | A distinct outer Baillarger stripe is visible, but the inner stripe cannot be delineated from layer 6 due to high fibre density in sublayer 6a 1 Occurrence: superior frontal gyrus, parts of precentral gyrus, medial orbital region, anterior cingulate cortex, inferior temporal gyrus |
|
| Unitostriate (conjunctostriate) | Both Baillarger stripes appear to be fused to a broad band, due to an increased fibre density in layer 5a (accompanied by a more or less prominent thinning of this layer) and a lightly stained layer 6a Occurrence: lateral orbital region, posterior part of the inferior frontal gyrus, parts of extrastriate visual cortex |
|
| Singulostriate | The inner Baillarger stripe is lacking Occurrence: temporopolar region, primary visual cortex |
|
| Myelin density in Baillarger stripes | ||
| Equodensus | Both Baillarger stripes are equally dense | |
| Externodensior | Outer Baillarger stripe denser than inner stripe | |
| Internodensior | Inner Baillarger stripe denser than outer stripe | |
| Intrusion depth of radiate bundles into the cortical ribbon | ||
| Euradiate | Radiate bundles reach upper border of layer 3b | |
| Infraradiate | Radiate bundles reach upper border of layer 5b | |
| Supraradiate | Radiate bundles extend into layers 1–2 | |
The molecular layer I of the adult brain is cell-sparse and contains few scattered neurones. It consists mainly of the apical dendritic tufts of pyramidal neurones and horizontally orientated axons, as well as glial cells. Some spiny stellate, neurogliaform, horizontal and a few scattered Cajal–Retzius cells are found in this layer. Inputs to the apical tufts are thought to be of cortico-cortical origin, but thalamo-cortical axonal terminals also occur. There are several subtypes of Cajal–Retzius cells; they are numerous in the fetal brain, but most are eliminated during early postnatal stages. They release the neurotransmitter glutamate, but their precise function is not clear. They control the laminar organization of the developing cortex by releasing the glycoprotein Reelin which acts like a stop signal for the outwardly migrating neurones. A lack of Reelin or other molecules in its signalling pathway results in a severe impairment of laminar organization. Decreased Reelin has been associated with lissencephaly and major psychiatric disorders in the human brain. The horizontally running dendrites of Cajal–Retzius cells are embedded in a mass of tangentially running myelinated and unmyelinated axons and dendrites. These tangential fibres consist of afferent fibres that arise from outside the actual cortical area, fibres from cortical interneurones, and the apical dendritic arborizations of virtually all pyramidal neurones of layers III and V. In myelin stained sections, a narrow band of tangential fibres, the Cajal–Retzius stripe, is visible in the centre of layer I (see Fig. 32.9 ).
The external granular lamina II is dominated by the somata of small pyramidal and non-pyramidal cells (‘granular’ cells). Myelin staining shows mainly vertically orientated fibres traversing the layer. In many brain regions, tangentially orientated myelinated fibres are found in higher density at the border between layers II and III, the horizontal Exner stripe (see Fig. 32.9 ). This border is difficult to determine in those cortical regions where the neuronal density of layer II is low.
The external pyramidal layer III predominantly contains small and medium-size pyramidal cells, as well as non-pyramidal neurones with vertically orientated intracortical axons. Together with layers I and II, layer III is the main target of interhemispheric (callosal) cortico-cortical afferents, and the principal source of cortico-cortical efferents. Scattered interneurones are also found in this layer. The pyramidal cells are smallest in the most superficial part of the layer and their size increases towards its deeper part. Layer III can frequently be subdivided into IIIa, IIIb and IIIc, where IIIa is most superficial and IIIc is the deepest. As in layer II, myelin stains reveal a mostly vertical organization of fibres; a band of horizontally running and myelinated fibres, the outer Baillarger stripe, can be seen at the border between layer IIIc and IV in many isocortical regions (see Fig. 32.9 ).
The width of the internal granular layer IV varies according to area. In Nissl-stained preparations it is seen to consist mainly of small round granular cells, which are different types of spiny stellate and pyramidal neurones and various types of aspiny interneurones. Layer IV is the main target of thalamo-cortical afferents and intra-hemispheric cortico-cortical afferents. In myelin-stained sections, the outer band of Baillarger is seen in layer IV of the primary visual cortex. The band is so obvious in this cortical area that it can be detected with the naked eye in histological sections of unstained fresh or frozen brains. It was first observed by Francesco Gennari in the year 1776, and independently 10 years later, by Félix Vicq d’Azyr: the horizontal, myelin-rich band is, therefore, often named the stripe of Gennari or of Vicq d’Azyr. It subdivides layer IV of the primary visual cortex into three sublayers: IVa, IVb (where the stripe is found) and IVc.
The internal pyramidal layer V often contains the largest pyramidal cells of any cortical layer. However, actual sizes vary considerably from area to area; in some association areas, layer III pyramidal cells are larger than those of layer V (externopyramidization). Layer V pyramidal neurones give rise to axons that leave the cortex and run down to subcortical structures. Scattered non-pyramidal cells are also present in layer V. In many cortical areas, layer V can be subdivided into sublayers Va, Vb and Vc. In myelin stained sections, the lamina is seen to be traversed by ascending and descending vertical fibres and to contain another prominent band of horizontal fibres, the inner band of Baillarger (see Fig. 32.9 ).
The multiform layer VI consists of excitatory cells (pyramidal neurones with dendrites and axons that branch and terminate in layers IV to V, spiny stellate cells with local dendritic branches and axons that terminate in layer VI, inverted pyramidal neurones, and bipolar/fusiform and other odd-shaped cells), and inhibitory interneurones. The upper part of this layer (layer VIa) has a higher cell density than its lower part (layer VIb), which blends gradually with the underlying white matter. A clear demarcation of layer VIb from the white matter is difficult in most brain regions. In some regions the cells of layer VIb can cover an area several times thicker than the cortex.
Cortical layers and their connections
Non-thalamic subcortical afferents to the neocortex from the basal forebrain ( Chapter 1, Chapter 2, Chapter 3, Chapter 4 , cholinergic), pedunculopontine ( Ch5 , cholinergic) and laterodorsal tegmental ( Ch6 , cholinergic) nuclei, ventral tegmental area (A10, dopaminergic), raphe nuclei (B1–9, serotoninergic), and nucleus coeruleus (A6, noradrenergic), terminate throughout all cortical layers; the laminar pattern of the density of their endings varies considerably from area to area.
Thalamo-cortical afferent fibres are found particularly in layers III–IV, and to a lesser extent also in layers I–II. Cortico-cortical input projects as long-distance feedback input mainly to layers II–IIIa, and as short-distance feedforward connections to layers IIIb and IV. Cortico-cortical input from the contralateral hemisphere (callosal afferents) mainly ends in layer III. In general, the largest input to a cortical area tends to terminate mainly in layer IV. This pattern of termination is best seen in the major thalamic input to primary visual, auditory and somatic sensory areas ( Fig. 32.10 ).
Supragranular pyramidal cells, predominantly layer III but also lamina II, give rise to both association and commissural cortico-cortical efferents. Generally, short cortico-cortical fibres arise more superficially, and long cortico-cortical (both association and commissural) axons come from cells in the deeper parts of layer III. The cortico-cortical long-distance feedforward fibres originate in layers II–III, and the long-distance feedback fibres from layers V and VI. Layer V also gives rise to cortico-striate (to the basal ganglia), cortico-thalamic, cortico-tectal, cortico-bulbar/nuclear (including cortico-pontine) and cortico-spinal fibres. Layer VI is the major source of cortico-thalamic and cortico-claustral fibres (see Fig. 32.10 ).
Columns and modules
Experimental physiological and connectional studies have demonstrated an internal organization of the cortex that is at right-angles to the pial surface and the underlying layers, with vertical columns or modules running through the depth of the cortex ( Fig. 32.11 ). The term ‘column’ refers to the observation that all cells encountered by a microelectrode penetrating and passing perpendicularly through the cortex respond to a single peripheral stimulus, a phenomenon first identified in the somatosensory and visual cortices ( , , ). In the visual cortex, narrow (50 μm) vertical strips of neurones respond to a bar stimulus of the same orientation (orientation columns), and wider strips (500 μm) respond preferentially to stimuli detected by one eye (ocular dominance columns; see p. 531 ). Adjacent orientation columns aggregate within an ocular dominance column to form a hypercolumn, responding to all orientations of stimulus for both eyes for one point in the visual field. Similar functional columnar organization has been described in widespread areas of neocortex, including motor cortex and association areas.
The minicolumn is a distinctive module in the neocortex ( ). It is a highly repetitive cortical unit consisting of horizontal and vertical components. Minicolumns display a considerable structural and functional heterogeneity between areas. The potential impact of the minicolumn concept seems to be important for comparative neuroanatomy and brain pathology ( ).
Apical dendrites of layer V pyramidal cells form columnar structures called dendritic bundles ( ). Each bundle consists of several apical dendrites and extends through layer IV into layer III/II where the dendrites begin to ramify (see Fig. 32.11B ). Within the bundle, some dendrites approach each other so closely as to be separated only by the extracellular space. Dendritic bundles and minicolumns of neuronal somata are spatially closely related. The dendritic bundles lie between the columns of cell bodies. The transverse diameter of both (about 50 μm) is roughly the same in the entire isocortex. Neurones in a minicolumn can send their apical dendrite to different bundles, some of them through bifurcating apical dendrites; neurones in separate minicolumns can send their dendrites to the same bundle. They might correspond to narrower assemblies of neurones with radially invariant activation/response to motion in the visual, motor and auditory areas. Electrophysiological studies suggest that the vertical bundles of dendrites are the morphological substrate of vertical functional units. Myelinated axons entering or leaving the cerebral cortex are also not distributed at random but are organized in axonal bundles (see Fig. 32.11B ).
Maps of the Human Cerebral Cortex
The systematic study of variations in cortical layering across the cerebral mantle has produced several maps of the human and non-human brain. Currently, the most widely used map is the cytoarchitectonic map of which segregates the human iso- and allocortex into approximately 50 areas or fields ( Fig. 32.12 ). Brodmann’s map lacks data on the areas buried within the intrasulcal surface and on interindividual variability in the extent and location of the areas. The Brodmann–Jülich map is shown in Fig. 32.13 (with some examples of probabilistic maps in Fig. 32.14 ): functional specialization will be discussed in this chapter in the context of the cytoarchitectonic divisions of the cerebral lobes.
Similar criticisms apply to the more elaborate cytoarchitectonic map produced by ( Fig. 32.15 ): Tables 32.3 – 32.4 compare the parcellation in the maps of and . Ongoing work on a cytoarchitectonic map of the entire human cerebral cortex based on a quantitative, statistically testable parcellation method, and information on intersubject variability and the cortical areas within the intrasulcal surface, is presented here (see Fig. 32.14 ) in a map that is three-dimensional; the brain with the map can be tilted by the user into any direction of view.
TABLE 32.3
Comparison between the map and nomenclature system used by and those of , and the Jülich/Düsseldorf group
modified after
| , | , | Jülich/Düsseldorf Group | |
|---|---|---|---|
| Frontal lobe | |||
| BA4 | V42, V43 | 4a, 4p | |
| BA6 | V36–V41 | 6 (preliminary delineation) | |
| BA8 | Comparability not clear | n.d. | |
| BA9 | Comparability not clear | n.d. | |
| BA10 | V50, V51 | Fp1, Fp2 | |
| BA11, BA12 | V1, V4-V9, V10-V12 (ventral parts) | n.d. | |
| BA24 | Subregio typica (V17–V24, dorsal part of V12) | s24, p24a, p24b, pv24c, pd24cv, pd24cd | |
| Subregio medioradiata (V25–V32) | a24a′, a24b′, 24c′v 24c′d, p24a′, p24b′, p24dv, p24dd | ||
| BA25 | V13, V14 | 25a, 25p | |
| BA32 | V3, V10–V11 (dorsal parts), V33–V35 | s32, p32, 32′ | |
| BA33 | Subregio extrema (V15, V16) | 33 | |
| BA44 | V56, V57 | 44d, 44v | |
| BA45 | V58, V59 (dorsal part) | 45a, 45p | |
| BA46 | V53, V54 | n.d. | |
| BA47 | V59 (ventral part) | n.d. | |
| Parietal lobe | |||
| BA1 | V70 | 1 | |
| BA2 | V71, V72 (?) | 2 | |
| BA3 | V67 | 3a | |
| V69 | 3b | ||
| BA5 | V75 | 5L, 5M, 5Ci | |
| BA7 | V83, V85, V86, V87 (?) | 7A, 7P, 7PC, hIP3 | |
| BA23 | V77–V80,V 91–V96 | 23d, 23c, d23, v23 | |
| BA26 | Vλ areas except Vλ3 | n.d. | |
| BA29 | Vλ3 | 29l, 29m | |
| BA30 | Vλ areas except Vλ3 | 30 | |
| BA31 | V76, V81, V82, V84 | 31 | |
| BA39 | V90 | PGa, PGp | |
| BA40 | V73-V74 (parts), V88, V89 | OP 1, OP2, PFop, PFt, PFcm, PF, PFm | |
| BA43 | V68, V72 (?), V73-V74 (parts) | OP 3, OP 4 | |
| Occipital lobe | |||
| BA17 | Area striata | hOc1 | |
| BA18 | Area occipitalis | hOc2 | |
| BA19 | Area praeoccipitalis | hOc3d, hOc3v, hOc4d, hOc4v, hOc5 (areas located within the region of BA19 and lateral BA37) and other areas. Note that a homogeneous cytoarchitectonic or functional area 19 does not exist | |
| Temporal lobe | |||
| BA20 | n.d. | n.d. | |
| BA21 | n.d. | n.d. | |
| BA22 | n.d. | Te3, Te4 (areas within BA22) | |
| BA38 | V38 | n.d. | |
| BA41 | Area temporalis transversa interna | Te1 (Te1.0, Te1.1, Te1.2) | |
| BA42 | Area temporalis transversa externa | Te2 (Te2.1, Te2.2) | |
| BA37 | n.d. | Fg1, Fg2 (on the fusiform gyrus as parts of medial PH) | |
| Insular lobe | |||
| Regio insularis anterior and posterior | Vi 1 –Vi 6 , Vai 1 –Vai 4 , Vai 6 , Vai 7 | Ig1 and Ig2 (dorsal part of Vi 5 ), Id1 (Vai 6 ). Most of the dysgranular and agranular areas have not yet been parcellated |
|
| n.d. | VTb 1 , VTb 2 | n.d. | |
Abbbreviation: n.d., not determined.
The Vogt map is mainly based on myeloarchitectonic observations, but also includes comparisons with cytoarchitecture. The other maps are based on cytoarchitectonic observations.
The Düsseldorf/Jülich map is based on the following publications: , , , , , , , , , , , , , , , , , .
Maps of cortical segregation using myeloarchitectonic criteria contained over 100 myeloarchitectonic areas ( Figs 32.16 – 32.18 ) ( , ) ( Fig. 32.19 ; see Table 32.2 ). The various cyto- and myelo-architectonic maps that were produced using different staining methods were roughly comparable: one of Brodmann’s areas ( ) contains several subdivisions represented in the maps of and of and in the probability maps ( Tables 32.3 – 32.4 ). However, some of the areas in Brodmann’s map (e.g. BA19, BA37), particularly in the higher eulaminate regions, could not be corroborated using more observer-independent methods and comparison with functional observations ( ).
TABLE 32.4
List of cytoarchitectonic areas according to (1) , (2) and (3) JuBrain ( ). The areas listed under Brodmann and JuBrain overlap completely or partially with those of von Economo and Koskinas. In (1), main areas (bold font) are distinguished from their subdivisions (normal font) as originally proposed by von Economo and Koskinas.
| 1 | Area name after von Economo and Koskinas | 2 | 3 | |
|---|---|---|---|---|
| Frontal lobe (F) | ||||
| Prerolandic region | ||||
| 1 | FA | Precentral area | 4 | 4a, 4p |
| 2 | FAγ | Giant pyramidal precentral area | 4 | 4a, 4p |
| 3 | FAop | Opercular precentral area | 4 | 4a, 4p |
| 4 | FB | Agranular frontal area | 6 | 6 |
| 5 | FBop | Opercular agranular frontal area | 6 | 6 |
| 6 | FC | Intermediate frontal area | 8 | n.d. |
| 7 | FCL | Limbic intermediate frontal area | 32 | n.d. |
| 8 | FCBm | Magnocellular agranular intermediate frontal (Broca’s) area | 44 | 44 |
| 9 | FCI | Intermediate frontal area at beginning of insula | n.d. | n.d. |
| 10 | FCDop | Opercular intermediate granular frontal area | n.d. | n.d |
| Anterior frontal (prefrontal) region | ||||
| 11 | FD | Granular frontal area | 9, 46 | n.d. |
| 12 | FDm | Magnocellular granular frontal area | 9 | n.d. |
| 13 | FDp | Parvocellular granular frontal area | 9 | n.d. |
| 14 | FDop | Opercular granular frontal area | 46 | n.d. |
| 15 | FDL | Limbic granular frontal area | 32 | p32 |
| 16 | FDΔ | Middle granular frontal area | 46 | n.d. |
| 17 | FDΓ | Triangular granular frontal area | 45 | 45 |
| 18 | FE | Frontopolar area | 10 | Fp1, Fp2 |
| 19 | FEL | Limbic frontopolar area | 32 | p32 |
| Orbital (orbitomedial) region | ||||
| 20 | FF | Granular orbital area | 47 | Fo3 |
| 21 | FFα | Agranular orbital area | 47 | Fo3 |
| 22 | FFΦ | Pretriangular orbital area | 47 | n.d. |
| 23 | FG | Area of straight gyrus (area recta) | 11 | Fo1 |
| 24 | FGi | Internal area of straight gyrus | 11 | Fo1 |
| 25 | FH | Prefrontal area | 11, 12, 32 | Fo2, n.d., s32 |
| 26 | FHL | Parolfactory prefrontal area | 11 | |
| 27 | FH L | Limbic prefrontal area | 32 | s32 |
| 28 | FJ | Frontoinsular area | n.d. | n.d. |
| 29 | FK | Frontal piriform area | n.d. | n.d. |
| 30 | FL1 | Primary parolfactory area | 25 | 25 |
| 31 | FL2 | Secondary parolfactory area | 25 | 25 |
| 32 | FL3 | Tertiary parolfactory area | 25 | 25 |
| 33 | FM | Geniculate area | 25 | 25 |
| 34 | FMt | Geniculate area of olfactory triangle | n.d. | n.d. |
| 35 | FN | Precommissural area | n.d. | n.d. |
| Superior limbic lobe (L) | ||||
| Anterior superior limbic region | ||||
| 36 | LA1 | Precingulate agranular anterior limbic area | 24 | p24c |
| 37 | LA2 | Anterior cingulate agranular anterior limbic area | 24 | s24b, p24b |
| 38 | LA3 | Cingulate agranular anterior limbic area limitans | 24 | s24a, p24a |
| 39 | LB1 | Anterior ultracingulate area | 33 | 33 |
| 40 | LB2 | Area of indusium griseum | Indusium griseum | n.d. |
| Posterior superior limbic region | ||||
| 41 | LC1 | Dorsal posterior cingulate area | 31 | n.d. |
| 42 | LC2 | Ventral posterior cingulate area | 23 | n.d. |
| 43 | LC3 | Posterior cingulate area limitans | 23 | n.d. |
| Retrosplenial subregion | ||||
| 44 | LD | Agranular retrosplenial area | 30 | n.d. |
| 45 | LE1 | Superior retrosplenial area granulosa | 29 | n.d. |
| 46 | LE2 | Inferior retrosplenial area granulosa | 29 | n.d. |
| 47 | LF1 | Posterior ultracingulate area | 26 | n.d. |
| 48 | LF2 | Ultracingulate area obtecta | 26 | n.d. |
| Insular lobe (I) | ||||
| 49 | IA1 | Dorsal precentral insular area | n.d. | n.d. |
| 50 | IA2 | Ventral precentral insular area | n.d. | n.d. |
| 51 | IB | Postcentral insular area | n.d. | Ig1, Ig2, Id1 |
| 52 | IBT | Postcentral insular area at temporal entrance | n.d. | n.d. |
| 53 | IC | Orbito-insular area | n.d. | n.d. |
| 54 | ID | Piriform insular area | n.d. | n.d. |
| Parietal lobe (P) | ||||
| Postcentral (anterior parietal) region | ||||
| 55 | PA1 | Giant pyramidal postcentral area | 3 | 3a |
| 56 | PA2 | Giant pyramidal postparacentral area | 5 | 5l |
| 57 | PB2 | Oral postcentral area simplex | 3 | 3b |
| 58 | PB1 | Oral postcentral area granulosa | 3 | 3b |
| 59 | PC | Intermediate postcentral area | 1 | 1 |
| 60 | PD | Caudal postcentral area | 2 | 2 |
| Superior parietal region | ||||
| 61 | PE(D) | Superior parietal area (transition postcentral) | ||
| 62 | PEm | Magnocellular superior parietal area | 7 | |
| 63 | PEp | Parvocellular superior parietal area | 7 | |
| 64 | PEγ | Giant pyramidal posterior superior parietal area | 7 | |
| Inferior parietal region | ||||
| 65 | PF | Supramarginal area | 40 | |
| 66 | PFt | Tenuicortical supramarginal area | 40 | PFt |
| 67 | PFop | Opercular supramarginal area | PFop | |
| 68 | PFcm | Magnocellular (posterior) supramarginal area | PFcm | |
| 69 | PG | Angular area | 39 | PGa, PGp |
| Basal parietal region | ||||
| 70 | PHP | Basal (temporo-occipital) parietal area at parietal entrance | 37 | |
| 71 | PHT | Basal (temporo-occipital) parietal area at temporal entrance | 37 | |
| 72 | PHO | Basal (temporo-occipital) parietal area at occipital entrance | 37 | |
| Occipital lobe (O) | ||||
| 73 | OA2 | Anterior peristriate area | 19 | |
| 74 | OA1 | Posterior peristriate area | 19 | |
| 75 | OAm | Magnocellular peristriate area | 19 | |
| 76 | OB | Parastriate area | 18 | hOC2 |
| 77 | OBγ | Giant pyramidal parastriate boundary of parastriate area | 17/18 border | |
| 78 | OBΩ | Maculae granulosae of parastriate area | 18 | |
| 79 | OC | Striate area (granulosa) | 17 | hOC1 |
| Temporal lobe (T) | ||||
| Supratemporal region | ||||
| 80 | TA1 | Posterior superior temporal area | 22 | |
| 81 | TA2 | Anterior superior temporal area | 22 | |
| 82 | TB | Magnocellular supratemporal area simplex | 42 | Te2 |
| 83 | TC | Supratemporal area granulosa | 41 | Te1 |
| 84 | TD | Intercalated supratemporal area | ||
| Temporal region proper | ||||
| 85 | TE1 | Middle temporal area proper | 21 | |
| 86 | TE2 | Inferior temporal area proper | 21 | |
| Fusiform region | ||||
| 87 | TF | Fusiform area | ||
| 88 | TH | Hippocampo-temporal area | 36 | |
| 89 | THα | Agranular hippocampo-temporal area | 35 | |
| Temporopolar region | ||||
| 90 | TG | Temporopolar area | 38 | |
| 91 | TGα | Agranular temporopolar area | 38 | |
| 92 | TJ | Temporal piriform area | ||
| 93 | TK | Area of substantia perforata | ||
| Hippocampal (inferior limbic) lobe (H) | ||||
| 94 | HA1 | Primary uncinate area | 34 | |
| 95 | HA2 | Secondary uncinate area | 34 | |
| 96 | HA3 | Tertiary uncinate area | 34 | |
| 97 | HB1 | Primary para-uncinate area | 28 | |
| 98 | HB2 | Secondary para-uncinate area | 28 | |
| 99 | HC | Rhinal area limitans | 28 | |
| 100 | HD1 | Presubicular area granulosa limitans | 27 | |
| 101 | HD2 | Middle presubicular area granulosa | 27 | |
| 102 | HD3 | Glomerular presubicular area granulosa | 27 | |
| 103 | HE1α | Subicular glomerular pyramidal area | ||
| 104 | HE1β | Subicular pyramidal area simplex | ||
| 105 | HE2 | Pyramidal area of Ammon’s horn | ||
| 106 | HE3 | Pyramidal area of digitate gyrus of uncus | ||
| 107 | HF | Dentate area | ||
The development of in vitro tissue slice binding techniques and quantitative autoradiography has enabled the visualization and quantification of neurotransmitters and their receptors within individual cortical fields. Although a single receptor type can characterize the borders of functionally and cytoarchitectonically defined cortical regions, in most cases only the characterization of the expression patterns of multiple receptors in the same cortical region can provide the criteria for a multimodal parcellation. These multi-receptor expression patterns are the characteristic fingerprints of each cortical area ( Figs 32.20 – 32.22 ). They facilitate detailed mapping of the human cerebral cortex and provide insight into the hierarchical relationships of these areas and the principles of their organization into structurally and functionally defined networks ( , , , , ). Maps of cortical neurotransmitters are important for understanding the neuropathological signatures of neurological and psychiatric disorders and the development of new molecules that can selectively modulate the activity of individual areas.
Transmitters in the Human Cerebral Cortex
A single receptor type can characterize the borders of functionally and cytoarchitectonically defined cortical regions. Prominent examples are the M 2 receptor and the delineation of BA17, area 3 and BA41 from surrounding cortical areas, or the identification of mossy fibre terminations on kainate receptors in the hippocampus. However, in many cases only the comparative evaluation of the expression patterns of multiple receptors in the same cortical region provides the criteria for a multimodal parcellation, which cannot be achieved by a single receptor type analysis. These multireceptor expression patterns, the characteristic fingerprints of each cortical area (see Figs 32.20 – 32.22 ), allow a detailed mapping of the human cerebral cortex and the identification of cortical networks. The analysis of receptor fingerprints reveals cortical regions that belong to the same functional network. For example, analysis of receptor fingerprints in the cingulate cortex has revealed four regions that are differentially involved in emotional processing, mediation of motor and cognitive processes via premotor planning with motivational characteristics, spatial orientation, memory and visuospatial functions ( ).
The fingerprints of BA44 and BA45 are more similar to each other than they are to those of the ventral premotor cortex or the ventrally adjacent BA47, often interpreted as part of Broca’s region. Subsequent analysis of many functionally identified language and non-language regions revealed that the receptor fingerprints of BA44 and BA45 belong to a complex system of language-related areas in the left hemisphere that consists of the subdivisions of BA44 and BA45, BA47, the secondary auditory area BA42, and a region in the posterior part of the superior temporal gyrus and sulcus within Wernick’s area ( , ; see Fig. 32.22 ). The fingerprints of the areas in this functional system clearly differ from those of primary sensory cortical areas, multimodal areas of the inferior and superior parietal lobules, higher extrastriate visual areas, BA32 and the primary motor cortex. A separate analysis of the fingerprints of the right and left hemispheres revealed a notable interhemispheric difference that underlines the lateralized language function.
Fingerprint analysis of the different areas of the inferior parietal lobule demonstrated a very close relation between BA39 and the extrastriate area V3v ( ), supporting the notion that at least the posterior part of BA39 is a hub in the ventrodorsal visual stream ( ). Hierarchical cluster analysis of the receptor fingerprints of fusiform areas FG1 and FG2 argues for their position between the ventral visual stream and multimodal association areas of the inferior parietal lobule ( ). FG1 and FG2 differ from the early visual areas of the ventral stream: a principal component analysis found that all visual areas were separated from nearly all other cortical areas studied, but FG1 and FG2 were segregated from early visual areas, indicating their close association with multimodal association areas of the inferior parietal lobule ( ).
Glutamate
Glutamate is the ubiquitous major excitatory transmitter in the human cerebral cortex, found in pyramidal neurones and astrocytes. The latter rapidly take up synaptically released glutamate via specific transporter molecules and metabolize it into glutamine and water. Glutamine is discharged from astrocytes by transporters, enters nerve terminals by electrogenic transporters, and is converted back into glutamate which is used again for neuronal transmission or is assimilated into the neuronal Krebs cycle. This glutamate–glutamine shuttle is crucial for the function of glutamate: excessive activation by high glutamate concentrations leads to pathological conditions and cell death.
Numerous glutamatergic pathways can be identified in the forebrain (see Fig. 32.7 ; Figs 32.23 – 32.24 ). Axons of glutamatergic pyramidal neurones in layers II and III project to cortical neurones of layers IV (short distance cortico-cortical feedforward projections) and V–VI (long distance cortico-cortical feedforward projections), and reciprocally from layers V and VI to layer II and superficial part of layer III (feedback projections): these are cortico-cortical pathways. Axons in layer VI project to the claustrum (cortico-claustral pathway). Axons in layer V excite GABAergic medium-spiny neurones (principal neurones) in the caudate nucleus and putamen via asymmetric synapses on their dendritic spines (cortico-striatal pathway). Axons in layers V and VI are connected via the internal capsule with neurones in the thalamus (cortico-thalamic pathway). Axons of neurones in thalamic nuclei mainly terminate on neurones in layer IV (thalamo-cortical pathway) and can extend to various degrees into layer III. Axons in layer V terminate on dopaminergic and GABAergic neurones in the substantia nigra (SN)/ventral tegmental area (VTA) (cortico-nigral/VTA pathway). Axons in layer V descend via the internal capsule and crus cerebri to multipolar cells of the pontine nuclei (cortico-pontine pathway), which in turn project as mossy fibres to the cerebellar cortex. Axons in layer V terminate via the internal capsule and crus cerebri on multipolar cells of the motor nuclei of the brainstem and tectum (cortico-nuclear and cortico-tectal pathways). Axons in layer V terminate via capsula interna, crus cerebri and pyramidal tract on motoneurones of the anterior horn of the spinal cord (cortico-spinal pathway). Axons in layer III terminate as commissural fibres via the corpus callosum on neurones of homologous or heterologous cortical areas of the contralateral hemisphere (callosal pathway).
GABA
GABA is the ubiquitous major inhibitory neurotransmitter in the human cerebral cortex. However, it can also have an excitatory effect under certain conditions. GABA-induced depolarization occurs in the developing cortex, and in mature cortical neurones following dendritic GABA A receptor activation. All cortical interneurones, with the notable exception of the spiny stellate cells, are GABAergic. Ninety-five per cent of cortical interneurones co-express GABA and calcium-binding proteins (CBPs): parvalbumin (PV), calbindin (CB), and calretinin (CR). Most of these cells are located in layers II–III. In the primary visual cortex of the macaque monkey, the majority of interneurones (97%) express only one of the three CBPs ( Fig. 32.25 ): about half of the cells are positive for PV, nearly one-third for CB and the rest are positive for CR. The three interneuronal subtypes also differentially express glutamate receptors ( ). The fast-spiking PV cells show high expression of the GluA2 and GluA3 subunits, whereas the intermediate spiking CB and CR cells show high expression of the GluA1 and GluA4 glutamatergic AMPA (alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid) receptor subunits ( ). Interneurones can be categorized according to their co-expression of the GluN2 subunits of the glutamatergic NMDA (N-methyl-D-aspartate) receptor. PV-IR cells show a low expression of the receptor, largely confined to their somata, whereas CB and CR expressing cells also express high levels of GluN2 subunits, both on their somata and in their dendrites ( ).
Monoaminergic neurotransmitters
The neuromodulatory transmitters of the cortex are acetylcholine, dopamine, noradrenaline and serotonin. Neurones that synthesize these neuromodulators are all located in subcortical regions and send long projection axons to the cerebral cortex. These transmitters can positively or negatively modulate the excitability of cortical neurones by binding to numerous different receptor subtypes linked to multiple signal transduction mechanisms: they enhance the signal-to-noise ratio of cortical responses and modify the threshold for activity-dependent synaptic modifications.
Acetylcholine is synthesized in large multipolar neurones in various regions of the basal forebrain ( Chapter 1, Chapter 2, Chapter 3, Chapter 4 )( Fig. 32.26 ). Their axons reach the entire cortex: cholinergic synapses are found on dendritic shafts, apical dendrites, the somata of pyramidal cell dendrites and of non-pyramidal cells. The density and laminar distribution pattern of cholinergic fibres differ significantly between different cytoarchitectonic areas. In the primary visual cortex, cholinergic fibres are particularly dense in layers I–IIIa, followed by layer IIIb, layers IVA and IVB, and alternating patches in layer IVC display a higher density of cholinergic fibres than layers IVB and V. Most but not all cholinergic synapses are symmetric.
Dopamine-containing neurones located in the dorsal parts of the substantia nigra (A9), ventral tegmental area (A10) and the retrorubral area (A8) send axons to the entire cortex (see Fig. 32.23 ). The whole frontal lobe receives dopaminergic afferents from these three sites ( ). Dopamine interneurones (A16) are present in the glomerular layer of the olfactory bulb. The densities and laminar distribution patterns of dopaminergic terminals differ between cortical areas: the primary motor cortex is densely innervated, the primary sensory cortices are sparsely innervated. In the motor cortex, dopaminergic fibres are evenly distributed throughout the different layers, other cortical regions show higher densities in layers I–IIIa and V–VI than in the middle layers. Dopaminergic terminals form symmetric synapses with dendritic shafts and spines, somata and axons. Direct local dopaminergic modulation of the excitability of cortical neurones is thought to be mediated via asymmetric membrane specializations of dopaminergic afferents on dendrites and dendritic spines, i.e. via the postsynaptic targets of both dopaminergic and excitatory input (synaptic triads) ( ).
Noradrenaline (norepinephrine) is synthesized in the locus coeruleus (see Fig. 32.23 ): the efferent projections reach the entire cortex via the medial forebrain bundle. Noradrenaline-containing axons are found in all cortical areas. There are differing regional and laminar densities, for example, in the human primary visual cortex, axon terminals are most dense in layer V, followed by layer VI, and relatively sparse in other layers. Noradrenaline is particularly important for the developmental plasticity of the cortex: it facilitates activity-dependent synaptic modifications in the cortex. Noradrenaline-containing varicosities form either symmetric or asymmetric synapses mainly on spines and dendritic shafts. Axosomatic synapses on pyramidal and non-pyramidal neurones are rare.
Serotonin (5-hydroxytryptamine, 5-HT) innervation of the entire cortex originates from either the dorsal raphe nucleus (see Fig. 32.23 ), via thin fibres with small varicosities, or the median raphe nucleus, via thick fibres with large spherical varicosities: the efferent projections from both raphe nuclei reach the entire cortex via the medial forebrain bundle. 5-HT axon terminals are present in all cortical areas and layers. 5-HT fibres synapse with pyramidal neurones and GABAergic interneurones, preferentially on spines and dendritic shafts, but also on somata.
Neuropeptides
Neuropeptides are small proteins or polypeptides that are synthesized by neurones and serve as modulatory neurotransmitters. They act in or out of the central nervous system via G protein and second messenger systems, and increase or decrease the functional impact of co-released neurotransmitters in the brain. More than 100 different neuropeptides have been identified, and only a selection of the most abundant types can be described here. Vasoactive intestinal polypeptide (VIP), neuropeptide Y (NPY), somatostatin (SOM), substance P (SP) and cholecystokinin (CCK) are among the frequently occurring neuropeptides in the human cerebral cortex. Many of the neuropeptide-containing neurones are interneurones. VIP is synthesized in bipolar interneurones and plays an important role in the local regulation of metabolism in the cerebral cortex by stimulating glycogenolysis and altering cortical blood flow. VIP inhibitory cells receive long range projections and inhibit SOM inhibitory interneurones. Dendrites of VIP cells most extensively branch in layers I and deep IV to superficial V. The density of their axons is highest in layers II–IV. NPY can be co-released with GABA and glutamate, reduces pain perception and voluntary alcohol intake, and controls epileptic seizures. Impairment of NPY functions also seems to play a role in obesity, anxiety, and neurodegenerative disorders. All SOM-expressing neurones of the cortex are a subset of GABAergic inhibitory interneurones (Martinotti cells, bitufted interneurones and basket cells). Basket cells are particularly abundant in the frontal or entorhinal cortices. SOM, NPY, SP and calbindin are often co-expressed in the same cell. The somata of such neurones occur mainly in the infragranular layers, or in the subcortical white matter. The activity of SOM cells is regulated by changes in brain state, during learning and reward: they undergo long-lasting structural and functional changes during experience-dependent plasticity of the cortical network. SOM cells play a role in schizophrenia and epilepsy. Only a few pyramidal cells of the isocortex contain SP, which is often co-localized with serotonin or noradrenaline. CCK can be co-localized with glutamatergic, GABAergic, dopaminergic and serotoninergic neurones that project to the cortex: it plays a role in regulating satiety and appetite, thermoregulation, sexual behaviour and memory. The cell bodies of CCK-containing neurones are frequently observed in the supragranular layers. Cell bodies of SP- and SOM-containing neurones prevail in the temporal lobe, those of CCK- and NPY-containing neurones in the frontal lobe. Axons containing NPY, SOM and SP are present in all cortical layers, particularly as horizontally orientated fibres in layer I, and randomly orientated fibres in the supra- and infragranular layers, but they are sparse in the inner granular layer IV.
Transmitter receptors in the human cerebral cortex
Transmitter receptors are key molecules of signal processing in the nervous system. The functional receptors are homomeric or heteromeric proteins in the cell membrane of neurones or glial cells ( Table 32.5 ). The effects of transmitters are defined by the receptors to which they bind after release from the presynaptic neurone. One transmitter can bind to multiple receptor types. This non-covalent binding leads to local changes of the membrane potential that may be fast (via ionotropic receptors, which are ligand-controlled ion channels) or slow (via metabotropic receptors, coupled to G-proteins that control the activation of second messengers). These changes are either excitatory or inhibitory depending on the actual receptor type. Receptors are located postsynaptically or presynaptically as heteroreceptors, i.e. receptors that regulate the synthesis and/or the release of transmitters other than their own transmitter, or as autoreceptors, i.e. receptors located on the terminal of a presynaptic cell. Autoreceptors are usually G protein-coupled receptors and only respond to the same transmitter released by the presynaptic neurone: they are located in the dendrites, soma, axon, or axon terminals, and act as part of a negative feedback loop in signal transduction.
TABLE 32.5
Receptor subtypes of the cerebral cortex and their subunit composition, genes, ion effects and function
| Transmitter and receptor type | Subunits | Genes | Ion | Function |
|---|---|---|---|---|
| Glutamate | ||||
| Subtypes | ||||
| Ionotropic AMPA receptor | GluA1-GluA4 | GRIA1-GRIA4 | Ca 2+ | Fast excitation |
| Ionotropic NMDA receptor | GluN1 | GRIN1 | Na + and Ca 2+ | Slow excitation |
| GluN2A-D | GRIN2A-D | - | ||
| GluN3A-B | GRIN3A-B | K + (out) | ||
| Ionotropic kainate receptor | GluR5-GluR7 | GRIK1-GRIK3 | Na + and Ca 2+ | Excitation (postsynaptic) |
| KA1, KA2 | GRIK4-GRIK5 | K + (out) | Inhibition (presynaptic) | |
| Metabotropic GluR2/3 receptor | GRM2, GRM3 | K + activation | Inhibit release of glutamate, GABA | |
| G-protein-coupled receptor | Ca 2+ inactivation | Noradrenaline, dopamine; pre-, post-, and extrasynaptic inhibition via cAMP inhibition | ||
| GABA | ||||
| Subtypes | ||||
| Ionotropic GABA A receptor | α 1 −α 6 , β 1 −β 3 | GABRA1-6 | Cl - (in) | Fast inhibition |
| γ 1 −γ 3 , δ, ε, π, | GABPB1−3 | α + γ provide benzodiazepine binding sites | ||
| θ | GABRC1-3 | |||
| Metabotropic GABA B | GABBR1-2 | K + | Slow inhibition G-protein-coupled receptor pre-, post-, and extrasynaptic | |
| Acetylcholinesterase | ||||
| Subtypes | ||||
| Ionotropic nicotinic α 4 β 2 receptor | Post- and presynaptic excitation | |||
| Metabotropic muscarinic M 1 receptor | Excitation | |||
| Metabotropic muscarinic M 2 receptor | Inhibition | |||
| Metabotropic muscarinic M 3 receptor | ? | |||
| Dopamine | ||||
| Subtypes | ||||
| Metabotropic D 1 receptor | Excitation | |||
| Metabotropic D 2 receptor | Inhibition | |||
| Noradrenaline (Norepinephrine) | ||||
| Subtypes | ||||
| Metabotropic α 1 receptor | Excitation | |||
| Metabotropic α 2 receptor | Inhibition | |||
| Serotonin (5-hydroxytryptamine) | ||||
| Subtypes | ||||
| Metabotropic 5-HT 1A receptor | Inhibition | |||
| Metabotropic 5-HT 2A receptor | Excitation/Inhibition | |||
The heterogeneous regional and laminar expression patterns of each receptor type indicate a parcellation of the cerebral cortex on a molecular basis. The regional segregation of transmitter receptors ( Fig. 32.27 ) can be demonstrated in vivo by positron emission tomography (PET) or single photon emission tomography (SPECT). Both methods are clinically important, but their spatial resolution does not enable an analysis of the laminar or cellular localization of receptors. In vitro demonstration of receptors at high spatial resolution and receptor type specificity is possible with immunohistochemistry, in situ hybridization or quantitative receptor autoradiography (see Fig. 32.27 ).
Glutamate receptors
Glutamate receptors are either ionotropic receptors (AMPA, alpha-amino-3-hydroxy-5-methyl-4-isoxazole-propionic acid; NMDA, N-methyl-D-aspartate; and kainate receptors), which control ion channels, or metabotropic receptors which act through second messenger systems. All glutamate receptor types are expressed by neurones and astrocytes. Activated AMPA and kainate receptors lead to fast excitatory synaptic signal transmission.
The highest densities of the ionotropic glutamate AMPA receptor ( Figs 32.28 – 32.29 ) are found in the orbitofrontal (BA11) and lateral prefrontal (BA8–10, 46, 47), temporal (BA20, 21, 36, 38), early and higher visual (V3A, FG1, FG2) and part of the insular cortical areas. Lowest densities are found in the motor (BA4, 6), somatosensory (BA1, area 3b), parietal (BA5, PGa, PGp), early visual (V2, V4v), and anterior and posterior cingulate (BA23, 31, 32) cortices. Preferential cellular localizations are the dendritic spines of pyramidal cells in the supragranular layers.
The ionotropic glutamate NMDA receptor ( Figs 32.28 – 32.29 ) is a heterotetrameric protein complex consisting of subunits NR1–NR3. It has a cation channel that enables sodium and calcium ions to move from extracellular space into a neurone after binding of glutamate and glycine. The current through the channel is voltage-dependent, since magnesium ions block the channel for the influx of sodium and calcium ions at resting membrane potential, i.e. the cell must be depolarized to remove magnesium ions from the channel before the influx of sodium and calcium ions can occur. Calcium ions are important for synaptic plasticity (long-term potentiation, LTP), a cellular mechanism for learning and memory. Since the effects of those synapses are potentiated, because the postsynaptic part is already depolarized by another channel, i.e. by AMPA receptors, the NMDA receptor is the molecular basis of Hebb synapses. Whereas LTP is caused by synaptic NMDA receptors, long-term depression (LTD) seems to be linked to extrasynaptic NMDA receptors. Excessive influx of calcium ions mainly via extrasynaptic NMDA receptors leads to excitotoxicity, found in Alzheimer’s, Parkinson’s and Huntington’s diseases, amyotrophic lateral sclerosis (ALS), stroke and epilepsy. Blocking of NMDA receptors by antagonists (e.g. memantine) is thought to be potentially helpful in the therapy of Alzheimer’s disease. Most NMDA receptors are preferentially expressed by pyramidal neurones and are localized postsynaptically on dendrites and dendritic spines in the supragranular layers II–III of the isocortex: their distribution is much sparser in layer IV. The glutamatergic thalamo-cortical input, which largely terminates in layer IV, is largely mediated by non-NMDA receptors. Relatively few NMDA receptors are presynaptic auto- and heteroreceptors. The highest NMDA receptor densities are found in parts of the insular, temporal (BA21, 36, 38), occipito-temporal (BA37), visual (V1, V2, V3v, V3A, V3d, FG1, FG2) and anterior cingulate (BA24, 32) cortices and the lowest densities are found in the motor (BA4, 6), part of the lateral prefrontal (BA9, 10, 46, 47), Broca’s (BA44, 45), parietal (BA5, PFt) and posterior cingulate (BA23) cortices.
The ionotropic kainate receptors (see Figs 32.28 – 32.29 ) are preferentially found in the anterior cingulate (BA24, 32), orbitofrontal (BA11), lateral prefrontal (BA8–10), somatosensory (BA2) and temporal (BA42, 20–22, 36) regions, and the lowest densities are found in the motor (BA4, 6), part of Broca’s region (BA45), superior parietal (BA5), occipito-temporal (BA37) and visual (V1, V2, V3A, Vd, V4v, FG1, FG) cortices. In the isocortex, kainate receptors are found with highest densities in the infragranular layers.
The metabotropic mGluR2/3 receptor (see Figs 32.28 – 32.29 ) reaches higher densities than any of the ionotropic glutamate receptors. The highest densities occur in the inferior temporal, anterior cingulate and entorhinal cortices. The vast majority of mGluR2/3 receptors occur as presynaptic autoreceptors in axonal terminals. The metabotropic mGluR1 receptor occurs only at very low concentrations in the cerebral cortex. The metabotropic mGluR5 receptor reaches highest densities in the cingulate and insular cortices.
GABA receptors
GABA receptors are either ionotropic or metabotropic. The highest densities of the inhibitory ionotropic GABA A receptor (see Figs 32.28 – 32.29 ) occur on somata and initial axonal segments in the supragranular layers of the primary and early visual cortex (V1, V2, V3v, V4v, V3d, V3A) followed by the primary somatosensory (area 3b), inferior parietal (PFm, PGp, PGp), posterior cingulate (BA31) and primary auditory (BA41) cortices. The lowest densities occur in the motor (BA4, 6), lateral prefrontal (BA8–10), anterior cingulate (BA24, 32), inferior parietal (PFt) and ectorhinal (BA36) areas, as well as part of Broca’s region (BA44).
The metabotropic GABA B receptor (see Figs 32.28 – 32.29 ) reaches generally higher densities than the GABA A receptor in the cerebral cortex. The highest densities are found in the orbitofrontal (BA11), lateral prefrontal (BA8–10, 47), primary somatosensory (BA2), superior parietal (BA5), anterior cingulate (BA24, 32), ectorhinal (BA36), insular and temporal (BA20, 21, 42) regions, and the lowest densities in the motor (BA4, 6) cortex, Broca’s region (BA44, 45), inferior parietal (PFm, PGa), temporo-polar (BA38), early visual (V2, V3d, V3A, V4v) and posterior cingulate (BA23) regions.
Monoaminergic receptors
Acetylcholine receptors are either ionotropic or metabotropic. The metabotropic muscarinic receptors (mAChRs) occur at considerably higher densities than the ionotropic nicotinic receptors (nAChRs). Five subtypes of mAChRs have been identified (M 1 –M 5 ). The excitatory M 1 , M 3 and M 5 receptors are coupled to Gq proteins which activate phospholipase C leading to influx of calcium ions and activation of intracellular signalling cascades. The inhibitory M 2 and M 4 receptors are coupled to G i/o proteins that inhibit the enzyme adenylyl cyclase and so reduce the production of cAMP.
The metabotropic excitatory M 1 receptor has seven transmembrane domains and suppresses potassium ion channels through G q protein coupling. The receptor mediates slow excitation. It reaches generally higher densities than the M 2 , but lower densities than the M 3 receptor (see Figs 32.28 – 32.29 ). The M 1 receptor plays a major role in learning and attention and the pathogenesis of amyloid plaques in Alzheimer’s disease: high receptor densities favour the non-amyloidogenic pathway in the metabolism of the amyloid precursor protein, whereas low expression of this receptor seems to favour the amyloidogenic pathway. Highest densities of M 1 receptors are found in the lateral prefrontal (BA46, 47), occipito-temporal (BA37), primary and secondary auditory (BA41, 42), temporal (BA21, 22), temporo-polar (BA38), ectorhinal (BA36), inferior parietal (PFm, PFt) and cingulate (BA24, 31) regions: nearly all of these regions are found in the default mode network. Motor (BA4, 6), lateral prefrontal (BA10), orbitofrontal (BA11), somatosensory (area 3a), early and higher visual (V2, V3A, V3d, V3v, FG1, FG2) and superior parietal (BA5) regions have the lowest densities.
The inhibitory M 2 and M 4 receptors have seven transmembrane domains, which act via G i/o proteins that lead finally to opening of potassium ion and closing of calcium ion channels. The M 2 receptor (see Figs 32.28 – 32.29 ) is located pre- and postsynaptically. Highest densities are found in layers III and IV of the primary sensory cortices, i.e. area 3b as part of the primary somatosensory cortex, BA41 (primary auditory cortex) and BA17 (primary visual cortex), where it improves the signal-to-noise ratio. Early visual areas (V2, V3v, V4v, V3d), Broca’s region (BA44, 45), lateral prefrontal (BA46), inferior parietal (PGa, PGp) and anterior cingulate (BA24, 32) areas also show high receptor densities, whereas the primary sensory areas of the isocortex, the motor cortex (BA4, 6), orbitofrontal (BA11), somatosensory (BA2), parietal (BA5, PFt), temporal (BA20–22), ectorhinal (BA36), occipito-temporal (BA37) and hippocampal areas show a distinctly lower M 2 receptor density.
The cholinergic excitatory metabotropic M 3 receptor also has seven transmembrane domains and acts via G q proteins. They upregulate phospholipase C and, therefore, inositol trisphosphate and intracellular calcium as a signalling pathway. Highest densities of this receptor (see Figs 32.28 – 32.29 ) are found in the supragranular layers of superior and inferior parietal (BA5, PGa, PGp, P, Pft), medial temporal (BA21), occipito-temporal (BA37), posterior cingulate (BA31), fronto-polar (BA10), lateral prefrontal (BA9, 46) and orbitofrontal (BA11) regions and lowest densities are seen in Broca’s region (BA44, 45), motor (BA4, 6), somatosensory (area 3a, BA1), temporo-polar (BA38), anterior cingulate (BA32) and visual (V2, V3A, V3d, V3v, V4v) cortices.
The neuronal nicotinic acetylcholine receptors consist of five subunits, of which each has four tansmembrane domains. The subunits are symmetrically arranged around the cation channel. Cations cause a depolarization: calcium ions act on different intracellular signal cascades that can lead to the gene regulation or release of transmitters. The nicotinic receptors are either homomeric or heteromeric (at least one α and one β subunit) combinations of 12 different subunits (α2−α10 and β2−β4). The excitatory nicotinic α 4 β 2 acetylcholine receptor (see Figs 32.28 – 32.29 ) reaches highest densities in the layer IV of the primary visual (V1), auditory (BA41) and somatosensory (areas 3a and 3b) cortices and layer III of the primary motor cortex (BA4). Relatively high densities are also found in the supragranular layers of the fronto-polar (BA10), orbitofrontal (BA11), lateral prefrontal (BA8, 9, 46, 47), cingulate (BA24, 23, 32), higher visual (FG1) and entorhinal (BA28) cortices. Lowest densities occur in the premotor (BA6), somatosensory (BA1, 2), inferior parietal (PFm, PFt), temporal (BA20–22), occipito-temporal (BA37), early visual (V2, V3d) and ectorhinal (BA36) areas.
Noradrenaline (norepinephrine) improves the signal-to-noise ratio in the processing of sensory stimuli and enhances consolidation of long-term memory. It is also important for regulating the working memory and attention functions of the prefrontal cortex. Moderate levels of noradrenaline enhance prefrontal cortical functions via post-synaptic α 2 adrenoceptors under normal conditions; high levels of noradrenaline during stress impair prefrontal cortical functions via α 1 receptors.
The excitatory metabotropic α 1 adrenoceptor (see Figs 32.28 – 32.29 ) is coupled to the G q protein, which causes an increase of intracellular calcium ions (via second messenger pathways), followed by a slow depolarization. Highest α 1 receptor densities are found in the supragranular layers of the cingulate (BA24, 31, 32), premotor (BA6), orbitofrontal (BA11), lateral prefrontal (BA8), somatosensory (BA2), inferior parietal (PFm), temporal (BA20–22), secondary visual (V2) and ectorhinal (BA 36) areas, whereas lowest values are found in the Broca region (BA44, 45) and the primary motor (BA4), somatosensory (area 3b, BA1), auditory (A41)and visual (V1) cortices. Low densities are also seen in posterior cingulate (BA23), inferior parietal (PFt) and early as well as higher visual areas (V3d, V3A, V4v, FG1, FG2).
Inhibitory metabotropic α 2 adrenoceptors (see Figs 32.28 – 32.29 ) is coupled to the G i protein. This causes a low cAMP level. The α 2 adrenoceptors are generally rarer than the α 1 adrenoceptors in the cerebral cortex. The α 2 adrenoceptors are found post- and presynaptically with highest densities in the supraganular layers. The presynaptic α 2 subtype occur as autoreceptors on noradrenaline-containing neurones and terminals and cause negative feedback on noradrenaline release, while both α subtypes are found post-synaptically. Like the cholinergic muscarinic M 2 receptor, by far the highest densities of the α 2 receptors are found in the primary sensory areas (V1, area 3a, 3b, 1, 2, BA41). This is in sharp contrast to the low densities of the expression of the α 1 receptor in these regions. High densities are also present in lateral prefrontal (BA8), anterior cingulate (BA24, 32), and early visual areas (V2, V3d, V3A, V3v). Lowest densities are found in the motor (BA4, 6), superior (BA5) and inferior (PFm, PFt, PGa) parietal, temporo-polar (BA38), occipito-temporal (BA37), higher visual (FG1, FG2), and posterior cingulate (BA23, 31) areas.
The excitatory metabotropic β 1 and β 2 adrenoceptors occur at lower densities than the α subtypes in the isocortex. They are coupled to a G protein, which causes high cAMP levels. Highest β adrenoceptor densities are found in the motor and layers III and IV of the temporal cortex, whereas lowest concentrations are located in the occipital pole.
Serotonin (5-hydroxytryptamine, 5-HT) modulates the excitability of cortical neurones by binding to 14 different receptor subtypes, mainly the 5-HT 1 A , 5-HT 1 B and 5-HT 2 subtypes. Serotonin influences sleep and arousal processes, executive functions of the prefrontal cortex, affective behaviour, aggression, learning and memory, and modulation of pain. 5-HT receptors are localized on GABAergic interneurones or projection neurones, and on axon terminals, dendrites and the somata of glutamatergic neurones as well as on cells that modulate noradrenergic, dopaminergic and cholinergic neurones.
The inhibitory metabotropic 5-HT 1 A receptor (see Figs 32.28 – 32.29 ) binds to G i proteins and thus inhibits adenylate cyclase, opens potassium channels, closes calcium channels and inhibits phosphatidylinositol (PI) turnover. This receptor is a somatodendritic autoreceptor on serotoninergic neurones of the raphe nuclei throughout the brainstem and modulates the pacemaker role of these neurones in a negative feedback loop. 5-HT 1 A receptors are predominantly found in layers I to upper III of the isocortex, localized mainly in somata and dendrites of pyramidal cells in the most superficial layers, and at a much lower level on the initial axonal segment of layer V and VI pyramidal cells. The highest densities of these receptors are seen in the most posterior part of lateral prefrontal (BA46, 47), orbitofrontal (BA11), somatosensory (BA2), inferior parietal (PFt), secondary auditory (BA42, auditory belt), temporal (BA20–22), temporo-polar (BA38), occipito-temporal (BA37), anterior cingulate (BA24), ecto- (BA36) and entorhinal (BA28) cortices. The lowest densities are found in motor (BA4, 6), primary somatosensory (area 3b) and visual (V1) regions, as well as the BA44 in the Broca region, PGa in the inferior parietal, early and higher visual (V2, V3d, V3A, V3v, FG1, FG2), and posterior cingulate (BA 23) areas.
The inhibitory metabotropic 5-HT 1 B receptor is almost always postsynaptically localized. It is found at high density in the primary visual cortex.
The excitatory metabotropic 5-HT 2 receptors are coupled to Gq proteins; their activiation ultimately leads to an increase in the concentration of intracellular calcium ions and PI hydrolysis. The highest densities of the 5-HT 2 receptor (see Figs 32.28 – 32.29 ) are found in layers III and IV of primary sensory areas (area 3b, BA2, BA17, BA41), posterior parietal (BA5, PGa), secondary auditory (BA42), early visual (V2, V3v), middle temporal (BA21), temporo-polar (BA38), early visual (V3v), and anterior and posterior cingulate (BA24, 23, 31) regions. The lowest concentrations have been reported in the motor cortex (BA4, 6), Broca’s region (BA44, 45), lateral prefrontal (BA8–10), somatosensory (BA1), inferior parietal (PGp), occipito-temporal (BA37) and higher visual (FG1 and FG2) regions. 5-HT 2 A receptors play a major role in the psychotomimetic effects of serotoninergic hallucinogens, e.g. LSD; they are localized postsynaptically on layer II, III and V pyramidal cells, GABAergic interneurones and glial cells.
5-HT 3 , 5-HT 4 and 5-HT 7 receptors are present at very low densities in the cerebral cortex. The 5-HT 3 receptor, the only ionotropic 5-HT receptor, is a ligand-gated sodium channel, found in calbindin-, calretinin- and CCK-containing interneurones. The excitatory metabotropic 5-HT 4 and 5-HT 7 receptors are coupled to G s proteins and activate adenylate cyclase. In the isocortex, the 5-HT 4 receptor occurs at low levels and the highest densities of the 5-HT 7 receptor occur in the infragranular layers.
Dopamine receptors are represented by five metabotropic receptor subtypes each with seven transmembrane domains. The two D1-like receptor subtypes (D 1 and D 5 ) are coupled to the adenylate cyclase-activating G s protein, and the three D2-like receptors (D 2 , D 3 , and D 4 ) are G i protein-coupled receptors that inhibit adenylyl cyclase and activate potassium channels. Dopamine receptors are involved in locomotion, cognition, emotion and affect as well as neuroendocrine secretion. Both D1-like receptor subtypes are expressed in layer V pyramidal neurones of the prefrontal, premotor, cingulate and entorhinal cortex, and are localized pre- and postsynaptically on dendritic spines and shafts. Highest densities of the excitatory D 1 receptor (see Figs 32.28 – 32.29 ) are present in the primary somatosensory (area 3b, BA2), fronto-orbital (BA11), inferior parietal (PGp), occipito-temporal (BA37), primary visual (V1), early visual (V2, V3v, V3d, V3A, V4v), temporopolar (BA38) and anterior cingulate (BA24) regions. Lowest densities are found in the motor cortex (BA4, 6), Broca’s region (BA44, 45), somatosensory (area 3a), inferior parietal (PGa), temporal (BA20–22, 42), ectorhinal (BA36) and higher visual (FG 1, FG2) regions.
D 2 receptors are present at very low density in the deeper layers of the entire iso- and allocortex. The relatively highest D 2 densities are found in the prefrontal and temporal regions, and the glomerular and internal plexiform layers of the olfactory bulb, hippocampus and amygdala.
D 4 receptors are present in the hippocampus and isocortical regions in pyramidal cells and GABAergic interneurones.
Fundamental Segregation of Cortical Structures
The two major architectonic types in the cerebral cortex are the isocortex and the allocortex. Isocortex is also called neocortex because it developed late during brain evolution, coinciding with the appearance of the first mammalian species. All areas of isocortex have a six-layered structure that emerges during the late fetal period. Isocortex forms the largest part of the human cortex ( Fig. 32.30 ). It is described as homogenetic because of its uniform pattern of lamination ( ): adult isocortex is either homotypical (eulaminate) or heterotypical (koniocortical or dysgranular/agranular). Allocortex is heterogenetic ( ) and is distinguished developmentally by the fact that it does not develop through the cortical plate. In general, allocortex is composed of fewer layers than the isocortex (e.g. in the hippocampus and olfactory cortex), although the entorhinal cortex has more than six layers. Allocortex is subdivided into paleocortex and archicortex (hippocampus and part of the cingulate cortex). Transition zones, where the number and thickness of cortical layers vary, occur between the allo- and isocortex and are called peripaleo-, periarchi- and proisocortex, depending on the adjoining parts of allo- and isocortex ( Fig. 32.31 ).
Eulaminate isocortex retains the six-layered structure with a well-defined layer IV and is found in most of the prefrontal, parietal, temporal and occipital cortices. Heterotypical koniocortical cortex is found in the primary somatosensory (area 3b), visual (BA17) and auditory (BA41) areas that are characterized by the presence of layer IV granule cells in the more superficial layers (granularization). Heterotypical dysgranular/agranular cortex is observed in cortical areas where layer IV appears to be invaded by pyramidal cells from layers III and V to varying degrees, so that layer IV appears very thin and sometimes interrupted (dysgranular) or is not visible as a distinct layer (agranular) in Nissl-stained sections. The motor cortex (BA4 and BA6), prefrontal area BA8, anterior part of the insular lobe, posterior part of Broca’s language region (BA44), and the rostral area 32 of the anterior cingulate cortex have a dysgranular or agranular cytoarchitectonic appearance.
Paleo- and peripaleocortex
The paleocortex consists of the olfactory bulb, the area retrobulbaris (anterior olfactory nucleus), the tuberculum olfactorium and the area prepiriformis. The peripaleocortex is found in the most anterior part of the insular cortex adjacent to the paleocortex ( Fig. 32.31 ).
The olfactory nerves arise from olfactory receptor neurones in the olfactory mucosa. The axons collect into numerous small bundles ensheathed by glia and surrounded by layers of meninges and enter the anterior cranial fossa by passing through the foramina in the cribriform plate of the ethmoid bone. They are attached to the inferior surface of the olfactory bulb lying at the anterior end of the olfactory sulcus on the orbital surface of the frontal lobe, and terminate in the olfactory bulb.
Olfactory bulb
The olfactory bulb is continuous posteriorly with the olfactory tract, through which the output of the bulb projects directly to the ipsilateral retrobulbar area (anterior olfactory nucleus), olfactory tubercle, prepiriform area, amygdala and rostral entorhinal cortex. There is a clear laminar structure in the olfactory bulb ( Fig. 32.32A ). From the surface inwards, the laminae are the olfactory nerve layer, glomerular layer, external plexiform layer, mitral cell layer, internal plexiform layer and granule cell layer ( Fig. 32.32B ). The olfactory nerve layer is formed by the unmyelinated axons of the olfactory receptor neurones. These axons divide and synapse on terminal dendrites of secondary olfactory neurones (mitral, tufted and periglomerular cells) forming glomeruli. The external plexiform layer contains the principal and secondary dendrites of the mitral and tufted cells. The mitral cell layer is a thin sheet composed of the cell bodies of mitral cells, each of which sends a single principal dendrite to a glomerulus, secondary dendrites to the external plexiform layer and a single axon to the olfactory tract. It also contains a few granule cell bodies. The internal plexiform layer contains recurrent and deep axon collaterals of mitral and tufted cells and granule cell somata. The granule cell layer contains the majority of the granule cells and their superficial and deep processes; numerous centripetal and centrifugal fibres pass through the layer.
Different odour molecules are represented by different patterns of spatial activity in the olfactory bulb ( ). The axons of the principal neurones in the olfactory bulb (the mitral and tufted cells) form its output via the olfactory tract. These cells are morphologically similar and probably use glutamate or aspartate as their neurotransmitter. Mitral cells span the layers of the bulb and receive the sensory input at their glomerular tufts. The axons of mitral and tufted cells are parallel output pathways from the olfactory bulb. Periglomerular and granule cells are the main types of interneurones in the olfactory bulb. The majority of periglomerular cells are dopaminergic (cell group A15), but some are GABAergic. Granule cells are similar in size to periglomerular cells; they lack an axon and therefore resemble amacrine cells in the retina. Granule cells are GABAergic and have two principal spine-bearing dendrites that pass radially in the bulb. They are likely to be a powerful inhibitory influence on the output neurones of the olfactory bulb. D 1 receptors are present mainly in the internal granular and plexiform layers. The highest densities of D 2 receptors are found in the glomerular and internal plexiform layers of the olfactory bulb.
Neurones of the retrobulbar area and collaterals of pyramidal neurones in the olfactory cortex project to the granule cells of the olfactory bulb. Cholinergic neurones in the horizontal limb nucleus of the diagonal band of Broca, which is part of the basal forebrain cholinergic system, project to the granule cell layer and to the glomerular layer. Other afferents to the granule cell layer and the glomeruli arise from the noradrenergic pontine locus coeruleus and the serotonergic mesencephalic raphe nucleus.
Retrobulbar area
The granule cell layer of the olfactory bulb extends into the olfactory tract with scattered medium-sized multipolar neurones that constitute the retrobulbar area (see Fig. 32.31 ; 32.32C ). This area extends from the caudal part of the olfactory bulb to the point where the olfactory peduncle joins the cortex of the frontal lobe. Many centripetal axons from mitral and tufted cells relay to this area, and axons from the retrobulbar area join the remaining direct fibres from the bulb and continue into the olfactory tract.
As the olfactory tract approaches the anterior perforated substance, it flattens and splays out as the olfactory trigone. Fibres of the tract continue from the caudal angles of the trigone as diverging medial and lateral olfactory tracts that border the anterior perforated substance (see Fig. 32.31 ). The lateral olfactory tract follows the anterolateral margin of the anterior perforated substance to the insula, where it bends posteromedially to merge with an elevated region, the gyrus semilunaris, at the rostral margin of the uncus in the temporal lobe. The tenuous grey layer that covers the lateral olfactory tract merges laterally with the gyrus ambiens. The grey layer over the lateral olfactory tract and the gyrus ambiens form the prepiriform area, which extends caudally into the entorhinal area of the parahippocampal gyrus. The prepiriform area, periamygdaloid regions and the entorhinal area (BA28) together make up the piriform cortex. The anterior perforated substance is continuous laterally with the peduncle of the temporal stem and the amygdaloid complex, and medially with the septal region.
The main targets of the lateral olfactory tract are the prepiriform area, amygdala and rostral entorhinal cortex. Neurones in the prepiriform area project to targets that include the orbitofrontal and agranular insular cortex, thalamus (especially the medial dorsal thalamic nucleus), hypothalamus, amygdala and hippocampal formation.
Prepiriform area
The prepiriform area (BA51, primary olfactory cortex) is the largest olfactory cortical area and extends from the lateral olfactory tract to the limen insulae and temporal cortex ( Fig. 32.33 ). This paleocortical area can be subdivided into a frontal portion at the base of the frontal lobe, and a temporal portion in the region between the frontal and temporal lobes. It is partly underlain by a ventral extension of the claustrum (endopiriform nucleus) and shows the prototypical three-layered structure of the paleocortex, i.e. a superficial molecular layer, thin intermediate densocellular layer and a deep multiform layer. The superficial molecular layer is differentiated into three sublayers: an outermost, cell-poor sublayer (a), an intermediate sublayer (b) which contains many glial cells and the fibres of the lateral olfactory tract; and the deepest sublayer (c) which has a lower cell density. The intermediate densocellular layer varies in thickness and appears to be interrupted; it contains pyramidal and polymorphic cells with dendrites that extend in the first layer. The deep multiform layer consists of the largest cells in the prepiriform area with an intermediate packing density. A very thin, nearly cell-free sheet (external capsule) separates the multiform layer from the underlying claustrum.
The prepiriform area receives afferents from the olfactory bulb, retrobulbar area, olfactory tubercle, septum, amygdala, orbitofrontal cortex, and the contralateral prepiriform area (via the anterior commissure). Efferent reciprocal connections are found in all these areas, and in the anterior insular and hippocampal cortices, claustrum and ventral putamen.
Neuroimaging studies in humans demonstrate activity in the prepiriform cortex during olfactory learning and memory tasks as well as in olfactory-induced motivational and cognitive states: olfactory dysfunction is a major finding in anhedonia.
Olfactory tubercle
The olfactory tubercle is located caudal to the retrobulbar area and medial to the prepiriform area in the anterior perforated substance. Its three-layered structure is poorly differentiated in the human brain (see Figs 32.31 , 32.33C ). Numerous islands of Calleja, consisting of very small and densely packed granule cell neurones, are found in all layers of the tubercle and between the tubercle and the ventral striatum. The olfactory tubercle receives input from the olfactory part of the amygdala and from the olfactory bulb via the lateral olfactory tract and sends efferents to the ventral pallidum. Medium spiny cells are the dominant cell type; they resemble neurones in the striatum.
Periamygdaloid cortex
The periamygdaloid cortex (see Figs 32.31 , 32.33 ) is located ventral to the prepiriform area and medial to the entorhinal area. It is a cytoarchitectonically heterogeneous region with several subdivisions, all of which receive olfactory input from the olfactory bulb and other paleocortical regions. The periamygdaloid cortex is therefore considered part of the olfactory cortex.
Between the piriform cortex and the laterally adjoining insular cortex, a peripaleocortical transition zone (BA16, ) (see Fig. 32.31 ) can be demonstrated in non-human primate brains that has been called the peripaleocortical claustral region by , because of its close topographical relation to the claustrum. BA16 is the least differentiated area in the insular cortex and shows an agranular architecture. It seems to be comparable with the areas ID and TI of and shows a four-layered type (for details see ).
Connections of the peripaleocortical claustral region with the prepiriform cortex, amygdala, adjoining isocortex, claustrum and contralateral BA16 have been described in primates. The strong connections with the piriform cortex make an olfactory function of the peripaleocortical insular cortex probable, but a definite statement about the function of area 16 in the human brain is presently not possible.
You're Reading a Preview
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here
