Mechanisms of Radial Migration

Radial migration has classically been known as glial-guided cell migration because during this process neurons move along the processes of specialized glial cells known as radial glia (Rakic, 1971a, 1971b, 1972; Rakic et al., 1974; Edmondson and Hatten, 1987). Despite their name, however, radial glial cells do not simply function as static supportive elements. Instead, radial glial cells represent an intermediate stage in the stem cell lineage of the CNS (reviewed in Alvarez-Buylla et al., 2001) and undergo mitosis to produce new neurons (Noctor et al., 2001). In addition, radial glial cells have a process that spans the wall of the neural tube and reaches the pial surface (Bergman glial cells being one exception to this rule), where it is anchored to the basal membrane. This process establishes a point-to-point relation between the ventricular zone and the surface of the brain, supporting neuronal movement during radial migration. Genetic defects affecting the development of radial glia cells lead to abnormal neuronal migration in the CNS (reviewed in Ross and Walsh, 2001; Marin and Rubenstein, 2003), suggesting that radial glia integrity is fundamental for radial migration.

Although radial glia integrity is largely essential for radial migration, there seem to be exceptions to the rule described above. During early stages of corticogenesis, for example, new neurons undergo radial migration through a process known as somal translocation (described as perikaryal translocation by Morest, 1970), which appears to be largely independent of radial glial cells (reviewed in Nadarajah and Parnavelas, 2002). During somal translocation, the leading process of migrating cells terminates at the pial surface and it becomes progressively shorter as the cells approach their final position. This is also observed in cells moving through glial-guided radial migration as they approach the pial surface. Thus, for some cell types or specific developmental periods, radial migration may not directly depend on radial glial cells.

Radial migration has been preferentially studied during the development of the cerebral cortex and the cerebellum and thus most of our knowledge on the mechanisms that control radial migration derives from the analysis of these structures. In vitro and in vivo studies of radial cell migration have identified a number of molecules that mediate this mode of migration. These molecules belong to multiple categories, including motogenic factors (i.e., factors that promote migration), cell adhesion molecules, receptors, and secreted factors, some of which are described below. We have excluded from this list molecules controlling those aspects of migration that are likely to be common to any type of neuronal migration (e.g., LIS1, DCX; see Section 1.11.2), even though they have been classically associated with radial migration defects. In that context, it is worth noting that neuronal migration abnormalities are likely to be more easily identified in laminar than in nuclear structures; this does not exclude, however, a role for these molecules in other types of migration (see, for example, McManus et al., 2004b; Pancoast et al., 2005).

Several classes of molecules have been described to stimulate radial migration. In the cerebral cortex, for example, members of the neurotrophin family such as brain-derived neurotrophic factor (BDNF) and neurotrophin-4 (NT-4) promote the motility of cortical cells through their high-affinity receptor tyrosine kinase B (TrkB) (Behar et al., 1997; Brunstrom et al., 1997). Other factors, such as (GABA) and glutamate, also appear to promote the migration of cortical neurons in vitro. These neurotransmitters are released independently of the conventional soluble N-ethylmaleimide sensitive fusion protein attachment protein receptor (SNARE)-dependent mode of secretion - probably through a paracrine mechanism - and mediate their effects primarily through the activation of GABAA and N-methyl-D-aspartate (NMDA) receptors (Komuro and Rakic, 1993; Behar et al., 1996, 1999, 2001; Manent et al., 2005).

To a large extent, most factors directly involved in controlling radial migration are molecules that regulate the interaction between migrating neurons and radial glial. This is the case of Astrotactin-1 (Astn1), which was first identified as an activity mediating the interaction of neurons and radial glial processes in cerebellar cultures (Edmondson et al., 1988). Astn1 is a glycoprotein expressed by migrating neurons both in the cerebellum and in the cerebral cortex and it is required for normal migration of neuroblasts along glial processes (reviewed in Hatten, 2002). Integrins constitute another family of factors implicated in the association between migrating neurons and radial glia. Thus, function-blocking antibodies against a3, av, and p1 integrins perturb the interaction between neurons and radial glial cells in vitro (Anton et al., 1999; Dulabon et al., 2000). Moreover, radial migration is altered in the cerebral cortex of a3, a6, or p1 integrin mutant mice (Georges-Labouesse et al., 1998; Anton et al., 1999; GrausPorta et al., 2001), although the precise function of integrins during in vivo radial migration remains unsettled. In the case of a3 integrin, however, it has been suggested that signaling through this receptor may directly control actin dynamics and consequently influence the ability of migrating neurons to search and respond to guidance cues in the developing cortex (Schmid et al., 2004).

The interaction between migrating neurons and radial glial fibers may also be controlled through intra-cellular signaling cascades. For example, correct apposition of neurons to radial fibers may largely depend on the morphology of migrating neurons. In the cerebral cortex, migrating neurons are largely bipolar, which possibly facilitates their interaction with radial glial processes. In the absence of p35, a regulatory activator of Cdk5 that controls the function of many proteins associated with the cytoskeleton, the leading process of radially migrating neurons is branched, and this associates with an impaired neuro-nal-glia interaction and perturbed migration (Gupta et al., 2003). Thus, the morphological organization of migrating neurons might be an important factor in determining their mode of migration.

The interaction between migrating neurons and radial glial processes is important not only for the initiation and maintenance of radial migration, but also for the control of its finalization. The precise termination of radial migration is crucial for the normal organization of brain structures. This is more evident in cortical structures, in which the pattern of radial migration termination determines the establishment of the laminar organization. In the case of the isocortex, birth-dating studies have shown that layers in the cortical plate (future cortical layers 2-6) are established according to an inside-outside pattern, where the deeper layers contain cells that become postmitotic earlier than the cells in more superficial layers (Angevine and Sidman, 1961; Rakic, 1974). During development, new neurons migrate radially toward the surface of the cortex, passing through cohorts of previously born neurons, and detach from radial glia as they approach the marginal zone. Analysis of mutations in mice and humans has revealed that the interaction between migrating neurons and Cajal-Retzius cells, a specialized cell type present in the embryonic marginal zone, is essential for controlling the detachment of migrating neurons from radial glia and, subsequently, the normal laminar organization of the cortex (reviewed in Gupta et al., 2002; Marin and Rubenstein, 2003).

The interaction between Cajal-Retzius cells and radially migrating neurons is mediated, at least in part, by Reelin, a large glycoprotein secreted by Cajal-Retzius cells during early stages of the development of the cortex. Reelin is expressed in many regions of the developing brain and in many species of vertebrates, but its function has been most extensively studied in the developing cortex. (see Reelin, Cajal-Retzius Cells, and Cortical Evolution for more on Reelin and its function in neuronal migration.) Reelin is a high-affinity ligand for two members of the LDL family of lipoprotein receptors, the very low-density lipoprotein receptor (VLDLR) and the low-density lipoprotein receptor-related protein 8 (LRP8, also known as ApoER2), which are expressed by radially migrating cortical neurons (D'Arcangelo et al., 1999; Hiesberger et al., 1999). Signaling through VLDLR/LRP8 mediates tyrosine phosphorylation of the mouse homologue of the Drosophila protein Disabled (DAB1). DAB1 is a cytoplasmic adapter protein that interacts with the cytoplasmic tails of VLDLR and LRP8 and is linked to events related to the reorganization of the cytoskeleton.

Although many aspects of the function of the Reelin-VLDLR/ApoER2-Dab1 pathway in radial migration remain unsettled, it is clear that Reelin signaling is involved in the final events that lead to the detachment of migrating neurons from radial glia. Loss of Dab1 function, for example, results in an impairment of the adhesive properties of radially migrating neurons, which fail to detach normally from the glial fiber in the later stage of migration (Sanada et al., 2004). Importantly, the influence of Reelin on the adhesive properties of radially migrating neurons may be the result of its interaction with other proteins, such as a3p1 integrin receptors, which are expressed in radially migrating neurons (Dulabon et al., 2000). It should be noted, however, that Reelin function is likely not restricted to controlling the interaction between migrating neurons and glial fibers.

It is likely that the Reelin-VLDLR/ApoER2-Dab1 pathway is just one of many signaling routes controlling neuronal detachment from radial glia and movement termination in the cerebral cortex. Thus, other proteins that are specifically expressed in radial glial processes at the level of the cortical plate are also candidates for the regulation of this process. One of these proteins is secreted protein acidic and rich in cysteine-like 1 (SPARC-like 1), which appears to function in ending neuronal migration by reducing the adhesiveness of neurons to glial fibers in the cortical plate (Gongidi et al., 2004).

Was this article helpful?

0 0

Post a comment