Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
As progenitor cells proliferate and differentiate into neurons and glia (see Chapter 5 ), these cells migrate to form the cerebral cortex in an elaborate and still not completely understood series of genetically influenced processes. Neurodevelopmental studies and disorders of brain development have informed our knowledge of these processes. In fact, the period of early migration overlaps with neuronal proliferation, and the period of late migration overlaps with later cortical organization (see Chapter 7 ). The disorders referred to as disorders of migration are so termed if migration is thought to be the first or principal process of development affected.
Neuronal migration refers to the remarkable series of events whereby millions of neurons move from their sites of origin in the ventricular and subventricular zones to the loci within the central nervous system (CNS), where they will reside for life. The peak time period for this occurrence is the third to fifth months of gestation, although neuronal migration can be detected in certain areas of the cerebrum as early as the second month and after the fifth month (especially GABAergic interneurons; see later) ( Table 6.1 ). Regulation of the timing and direction of these many simultaneous migrations must be highly ordered, but only recently has insight been gained into these control mechanisms (see later).
Peak time period |
3–5 months |
Major events |
Cerebrum |
Radial migration: cerebral cortex (projection neurons), deep nuclei |
Tangential migration: cerebral cortex (interneurons) |
Cerebellum |
Radial migration: Purkinje cells, dentate nuclei |
Tangential migration: external → internal granule cells |
The major features of cell migration in the primate were defined initially, particularly by the classic studies of Sidman and Rakic, using primarily autoradiographic, electron microscopic, and Golgi techniques. Later work used immunocytochemical and retroviral methods and the study of genetically manipulated animals to elaborate earlier observations. Two basic varieties of cell migration have been delineated: radial and tangential (see Table 6.1 ). In the cerebrum , radial migration of cells from their origin in the ventricular and subventricular zones is the primary mechanism for formation of the cortex and deep nuclear structures. Radial migration gives rise to the projection neurons of the cortex. These neurons emanate primarily from the dorsal region of the subependymal germinative zones .
Tangential migration of neurons generated in the dorsal and ventral telencephalon, detailed in Chapter 5 , results in the gamma-aminobutyric acid (GABA)–expressing interneurons of the cerebral cortex. These neuronal precursors migrate parallel to the surface of the cortex and proceed in one of three streams (i.e., through the subventricular zone, the intermediate zone, or the marginal zone) before terminal radial movement to arrive in the cortical plate.
In the cerebellum , radial migration causes the genesis of Purkinje cells, the dentate nucleus, and other roof nuclei. Tangential migration of cells that originate in the germinal zones in the region of the rhombic lip and migrate over the surface of the cerebellum forms the well-known external granular layer . These cells then migrate radially inward to form the internal granule cell layer of the cerebellar cortex (see Chapter 4 ). Thus during their journey from their point of origin in the ventricular zone, the granule cells exhibit both radial and tangential migration.
The basic patterns of cell migration for formation of the cerebral cortex are shown in Figs. 6.1 and 6.2 . The first and earlier mechanism is movement by translocation of the cell body (i.e., somal translocation) ( Table 6.2 ). This mode of migration probably results in the formation of the preplate (see Fig. 6.1 ). This layer of neurons later is split by the arrival of the cortical plate neurons into a superficial layer nearest the pial surface, which produces the Cajal-Retzius and related neurons of the marginal zone, and a deeper layer, which becomes the subplate neurons. The preplate neurons and the subsequently formed Cajal-Retzius and subplate neurons are critical for the progression of neuronal migration (see later).
Initially, neurons migrate by translocation of the cell body (somal translocation). Later and predominantly, neurons migrate by following radial glial guides (i.e., the fibers of their clonally related radial progenitors). |
Simultaneous with this radial migration, tangential migration parallel to the surface of the cortex (followed by radial migration to the cortex) results in the placement of GABA-expressing interneurons throughout the cortex. |
Proliferative units of the ventricular zone migrate along the radial glial scaffolding to become the ontogenetic neuronal columns of cerebral cortex. |
Migration through subplate neurons and “waiting” thalamocortical and corticocortical afferents is likely important for later neuronal development (e.g., synaptogenesis). |
Early-arriving neurons take deep positions in cortex, and later-arriving neurons take superficial positions (i.e., inside-out pattern). |
The second mode of migration, leading to the formation of most of the cerebral cortex, occurs by radial migration (see Figs. 6.1 and 6.2 ). These cells are generated by the radial glial progenitors discussed earlier (see Chapter 5 ). The clonally related neuron migrates along the parental radial glial fiber, which extends to the pial surface. Initially, cells are generated in the ventricular zone , and then migrate relatively rapidly and synchronously through the intermediate zone in waves to the developing cortical plate. At later stages , as shown by Rakic in studies of the monkey visual cortex, the neurons are generated especially in the subventricular zone (see Fig. 6.1 ). By labeling dividing cells with [ 3 H]thymidine at various times during development and then determining where the labeled cells appear in the cortical plate, Rakic showed that cells that migrate first take the deepest positions in the cortex, whereas those migrating later take more superficial positions (“inside-out” pattern). By 20 to 24 weeks of gestation, the human cerebral cortex essentially has its full complement of excitatory and projection neurons, with neurons and glia having arrived from both radial and tangential migration streams and expressing various markers reflecting their neuronal identities, as schematized in Fig. 6.3 . A substantial proportion of GABAergic interneurons migrate in the latter months of gestation, such that by approximately term, GABAergic interneurons reach their peak density in the cortex.
How do the migrating cells know how to reach where they are going? In the major process of radial migration, radial glial cells serve as the guides for the migration of young neurons from their sites of origin in the ventricular and, later, subventricular zones, across a distance that can be many times greater than the length of their leading processes to their ultimate position in the cortical plate (see Table 6.2 and Fig. 6.4 ). The elaboration of the structure of the radial glial fiber system has been clarified by immunocytochemical and ultramicroscopic studies, especially by Caviness and others. Initially, the system is uniformly radial in alignment, and in the ventricular zone the fibers appear to separate columns of germinative cells—the proliferative units described by Rakic in the primate (see Chapter 5 ). The radial glial system in the developing cerebral wall forms fascicles of fibers rather than isolated fibers. With the rapid growth of the cerebral wall, and particularly the intermediate zone, the fiber fascicles develop distinct curves with definite region-specific changes in trajectory ( Fig. 6.5 ). Nevertheless, the dominant feature remains the migration of apparent clonally related columns of cells among the same radial glial fascicles, again likely related to the proliferative units described earlier. As the migrating neurons approach the cortical plate, the radial glial fascicles begin to defasciculate, and radial fibers tend to penetrate the cortex more as single fibers. This occurrence develops at the junction of the upper intermediate zone and the subplate zone, an important site for neuronal heterotopia in disorders of neuronal migration (see later). As outlined in Chapter 5 , it has long been postulated that the progeny of a single daughter cell arising from asymmetrical cell division at the ventricular zone gives rise to a column of neurons and that a columnar organization is thus established. Although this framework applies generally, there is also evidence that whereas some clonal populations of cells maintain regional specification, partially overlapping with other, neighboring clonal populations, other populations of neurons become widely dispersed through the cortex.
Insights into the key molecular determinants of neuronal migration have been gained in recent years ( Table 6.3 ). We present a brief discussion here of some of the best studied molecules; further review of molecular determinants is provided later in the discussions of the molecular aspects of individual disorders of migration. Roles for molecules on preplate neurons (and the later Cajal-Retzius and subplate neurons), radial glia, and migrating neurons have been established. From preplate neurons and the Cajal-Retzius cells of the marginal zone , such extracellular matrix molecules as fibronectin, chondroitin, and heparan sulfate proteoglycans are clearly crucial. The secreted glycoprotein, reelin, lacking in the mutant mouse “reeler” with a neuronal migrational disorder, is an important product of the Cajal-Retzius cells. Platelet-activating factor acetylhydrolase, lacking in one form of human lissencephaly (see later), suggests an important role for a molecule related to platelet-activating factor, a notion further supported by in vitro studies. Neurons with GABA receptors in the preplate, Cajal-Retzius cells, and perhaps migrating neurons also now appear to be involved in migrational events.
Preplate neurons (also marginal [Cajal-Retzius] zone | Fibronectin, chondroitin, and heparan sulfate |
and subplate neurons) and | Fukutin proteoglycans |
extracellular matrix | GABA receptors |
Integrins | |
Laminin | |
Reelin | |
Radial glia | erb B4 receptors Notch receptors |
BLBP | |
Migrating neurons | Neuregulin |
Astrotactin | |
Doublecortin | |
Platelet-activating factor acetylhydrolase (subunit 1) | |
Filamin 1 | |
Cyclin-dependent kinase-5 (cdk-5) | |
NCAM | |
NMDA receptors | |
Calcium channels | |
GABA receptors |
Important molecular determinants of migration on radial glia include three signaling pathways, involving erb B4 receptors, Notch receptors, and brain lipid-binding protein (see Table 6.3 ). The first of these are the surface ligands for neuregulin located on migrating neurons , which express several proteins involved in migration (see Table 6.3 ). An additional and well-characterized surface molecule of importance on migrating neurons is the glycoprotein astrotactin. Doublecortin is the product of a gene on the X chromosome and is involved in double cortex (band) heterotopia in females and lissencephaly in males, important human neuronal migration disorders (see later). This protein, expressed in migrating neurons at their growing end, is involved in an intracellular signaling pathway important for neuronal migration. Doublecortin is a microtubule-associated protein, which plays a role in microtubule polymerization and thereby may be involved in neuronal migration by mediating the cytoskeletal changes required for such movement, including through interactions with synaptic vesicle trafficking proteins. The gene filamin 1 (or FLNA ), which is responsible for the human neuronal migration disorder, periventricular heterotopia, encodes a neuronal actin cross-linking protein that transduces ligand-receptor binding into actin reorganization, critical for locomotion of many cell types. Involvement of neuronal calcium channels, glutamate, and N -methyl- d -aspartate (NMDA) receptors is supported by the work particularly of Rakic and colleagues but also of others. Thus selective blockage of N-type but not of L- and T-type calcium channels inhibits neuronal migration. N-type calcium channels are involved primarily in the release of neurotransmitters. That the neurotransmitter involved is glutamate, which acts on the NMDA receptor, is suggested by the demonstration that blockers of NMDA receptors, but not of non-NMDA receptors, inhibit neuronal migration ( Fig. 6.6 ). Axonal release of glutamate may be the source of the glutamate that acts on the migrating neuron.
Radial glial cells have additional functions aside from guidance of neuronal migration (see Chapter 5 ). Thus the initial role of these cells is as neuronal progenitors. Generation of neurons is followed by the role of radial glial guides. Radial glial cells also facilitate neurite (axon and dendrite) growth and subsequent connectivity. Still later, these cells give rise to astrocytes and oligodendroglia. As discussed in later chapters, the preterm brain is vulnerable to both white matter injury (premyelinating oligodendrocytes are especially vulnerable) and dysmaturation, with the cells and processes affected related to the timing of preterm birth and corresponding major periods in brain development ( Fig. 6.7 ). Finally, the cellular progeny of the radial glial cells then serve as a source of neural stem cells in the subventricular zone of the mature brain.
Disorders of neuronal migration usually cause overt disturbances of neurological function, with clinical deficits often apparent from the first days of life. Seizures are most often the dominant early neurological sign with the more severe migrational disturbances. The advent of magnetic resonance imaging (MRI) markedly increased the ability to identify these disorders in vivo, demonstrating the relatively high prevalence of these disorders and showing their broad clinical expression (see later). The major disorders are listed in Table 6.4 in order of decreasing severity. The disorders included here are considered migrational disorders conventionally, but as indicated, some have been found through molecular genetic studies to be due to dysfunction that begins during proliferation and continues to exact its major effects on migration and subsequently on cortical organization.
Order of decreasing severity |
Schizencephaly |
Agyria-pachygyria spectrum (e.g., lissencephaly) |
Polymicrogyria |
Heterotopia—periventricular, subcortical |
The hallmark of the migrational disorders is an aberration of gyral development. Formation of the many secondary and tertiary gyri of the human brain occurs after neuronal migration has ceased ( Fig. 6.8 ). The fastest increase in number of the major gyri occurs between 26 and 28 weeks of gestation. Further elaboration of these gyri continues during the third trimester and shortly after birth. The stimulus for gyral formation appears to be the remarkable increase in surface area of cerebral cortex that occurs during this period, particularly the difference in increase in surface area of the outer versus the inner cortical layers. In the normal cortex, the surface area of the outer cortical layers is greater than the inner layers, and this discrepancy leads to compressive stresses that may lead to gyral formation. These relative increases in cortical surface area require the complement of neurons provided by migrational events. In lissencephaly, in which all cortical layers fail to receive their full complement of neurons, no gyri develop. In polymicrogyria, the surface area of outer cortical regions is much greater than that of inner cortical regions, and the result is an excess of gyri. In addition, gyral development may be stimulated by the forces produced by growth of cerebral white matter axons originating in the cortex, the concept of tension-based morphogenesis (see Chapter 7 ).
In addition to gyral abnormality, a common feature of migrational disorders is hypoplasia or agenesis of the corpus callosum; occasionally, absence of the septum pellucidum also accompanies these disorders. Development of the corpus callosum (the major interhemispheric commissure) and of the septum pellucidum is associated temporally and causally with migrational events in the cerebrum (see Chapter 2 ). Thus the timing of these aspects of midline prosencephalic development is almost coincidental with the major neuronal migrational events for the formation of cerebral cortex. Moreover, normal elaboration of cortical callosal fibers requires normal progression of neuronal migration to cerebral cortex. The frequent concurrence of hypoplasia or agenesis of the corpus callosum with the migrational disorders discussed is therefore understandable. The preponderance of cortical abnormality at times dominates to the point that the callosal defect is not initially noted; in some settings, however, the callosal defect is a key feature of the abnormality—for example with ARX - and TUBA1A -associated lissencephaly, discussed later.
Schizencephaly is the most severe yet restricted of the cortical malformations (see Table 6.4 ). A complete agenesis of a portion of the germinative zones and thereby the cerebral wall is believed to exist, leaving seams or clefts. The pial-ependymal seam is characteristic ( Fig. 6.9 ). In the walls of the clefts, the cortical plate exhibits the hallmarks of migrational disturbance (i.e., a thick, microgyric cortex and large neuronal heterotopia). In bilateral lesions, schizencephaly in one hemisphere may be accompanied by polymicrogyria or focal cortical dysplasia (FCD) in the other (see later). The lips of the clefts may become widely separated, and dilation of the lateral ventricles may occur. Hydrocephalus often complicates such open-lipped lesions, especially when they are bilateral (see later). When they are bilateral, such open-lipped lesions may be referred to incorrectly as hydranencephaly, a later occurring destructive lesion of the cerebral hemispheres; when they are unilateral, they may be referred to incorrectly as porencephaly, a destructive lesion of one hemisphere. Indeed, it is now clear that unilateral and bilateral schizencephalies can be familial and probably account for previous reports of “familial porencephaly.” Gray matter (often polymicrogyric), lining the lesion and demonstrable on brain imaging, especially MRI, is the key finding indicative of schizencephaly.
The advent of MRI greatly expanded the understanding of anatomical and clinical aspects of schizencephaly (see Table 6.5 ). Indeed, several previous notions that were based almost exclusively on study of autopsy cases (i.e., that schizencephaly is rare, bilateral, and associated invariably with severe neurological deficits) were shown to be incorrect. Thus in two large series, 67 cases were collected, and schizencephaly was unilateral in 63% ( Table 6.5 ). The clefts tended to be in the regions of the rolandic and sylvian fissures and involved predominantly frontal areas. Subsequent series confirmed these observations.
FEATURE | PERCENTAGE |
---|---|
Anatomical | |
Unilateral | 63 |
Bilateral | 37 |
Closed clefts | 42 |
Open clefts | 58 |
Frontal | 44 |
Frontoparietal | 30 |
Parietal, temporal, or occipital | 26 |
Associated septo-optic dysplasia | 39 |
Anatomical-Clinical Correlates | |
Cognitive Disturbances (prominent) | |
Bilateral | 100 |
Unilateral | 24 |
Motor Disturbances | |
Bilateral | 86 |
Unilateral | 77 |
Frontal | 84 |
Not frontal | 29 |
Open lip | 94 |
Closed lip | 22 |
Seizure Disorder | |
Bilateral | 72 |
Unilateral | 60 |
Hydrocephalus | |
Open lip | 52 |
Closed lip | 0 |
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here