Jiy

Relative position of Relative position of centrosome, t = 0 leading process, t = 0

Figure 1 Steps in neuronal migration and the molecules involved. A prototypical migrating neuron contains distinct subcellular domains: the leading process, the perinuclear domain, and the trailing process. Neuronal migration involves repeated cycles of (1) polarized extension of the leading process, followed by movement of the centrosome forward, (2) a highly coordinated movement of the nucleus closer to the centrosome (nucleokinesis), and finally (3) a trailing process remodeling.

Relative position of Relative position of centrosome, t = 0 leading process, t = 0

Figure 1 Steps in neuronal migration and the molecules involved. A prototypical migrating neuron contains distinct subcellular domains: the leading process, the perinuclear domain, and the trailing process. Neuronal migration involves repeated cycles of (1) polarized extension of the leading process, followed by movement of the centrosome forward, (2) a highly coordinated movement of the nucleus closer to the centrosome (nucleokinesis), and finally (3) a trailing process remodeling.

and direct neuronal migration through the CNS, acting as chemorepellent factors for both migrating axons and neurons (Brose and Tessier-Lavigne, 2000) (see also Section 1.11.5).

During chemotaxis, migrating cells - including neurons - appear to detect very small differences in chemical gradients and therefore it is likely that the process of polarization requires their amplification through steeper intracellular gradients that allow appropriate cellular responses (Figure 1). In Dictyostelium cells, this process involves the polarization of phosphoinositides (such as phosphatidylinositol-triphosphate (PIP3) and phos-phatidylinositol (3,4)-biphosphate (PI(3,4)P2)P2) across the cell and is mediated by localized accumulation at the front of the cell of phospho-inositide 3-kinase (PI3K), which generates phosphoinositides, and restricted localization and activation at the rear of the phosphatase and Tensin homologue deleted on chromosome 10 (PTEN), which removes them (Funamoto et al., 2002). In neurons, however, very little is known about how these molecules control directed polarization. Nevertheless, PI3K is required for the chemotaxis of neurons in response to neurotrophins (Polleux et al., 2002) and perturbation of PTEN function causes abnormal neuronal migration (Li et al., 2003).

Directed migration also requires the polarization of several organelles in slow-moving cells such as neurons. Specifically, the microtubule-organizing center (MTOC) and the Golgi apparatus are normally localized ahead of the nucleus and plays a role in defining the direction of movement. (This is not the case for fast-moving cells such as neurotrophils, in which the MTOC is behind the nucleus.) In other cell types, the small Rho GTPase Cdc42 is active toward the front of migrating cells during chemo-tactic responses and plays a role in localizing the MTOC ahead of the nucleus, although its contribution to the polarization of migrating neurons is still unclear. Nevertheless, inactivation of Cdc42 appears to be required for Slit repulsion of migratory cells from the subventricular zone (SVZ) of the telencephalon (Wong et al., 2001), suggesting that Cdc42 may normally help to polarize migrating neurons toward a chemoattractant source but is inactivated during chemorepulsion. Another Rho GTPase, Rac, is also polarized to the front of migrating cells and is involved in promoting directional extension of protrusions through a signaling loop that involves also Cdc42 and PI3K products. In neurons, the cyclin-dependent kinase 5 (Cdk5) and its neuron-specific regulator p35 localize with Rac during the extension of neurites and are part of the signaling machinery that may help neurons to engage in directional migration (Nikolic et al., 1998). The direct interaction of Rac and possibly other small Rho GTPases with the cytoskeleton at the front of the cell appears to constitute the final effector mechanism that mediates the extension of migrating cells in a specific direction.

1.11.2.2 Nucleokinesis

One of the main differences that distinguish axon guidance from cell translocation is, obviously, the coordinated movement of the nucleus during cell migration. Thus, nucleokinesis is a fundamental step in the cycle that leads to directed cell migration and neurons are no exception to this rule. Indeed, disruption of nuclear translocation systematically leads to prominent defects in neuronal migration (Xie et al., 2003; Shu et al., 2004; Solecki et al., 2004; Tanaka et al., 2004).

Nucleokinesis in migrating neurons critically depends on the microtubule network, which plays a part in positioning the nucleus during translocation (Rivas and Hatten, 1995). As briefly mentioned in the previous section, polarization of neurons during migration includes the location of the MTOC ahead of the nucleus, an event that appears to be necessary for normal movement of the nucleus (Figure 1). This process relies on the interaction between the MTOC and the nucleus through a specialized network of perinuclear microtubules and microtubule-associated proteins, such as doublecor-tin (DCX) and lissencephaly-1 (LIS1). Both of these proteins bind to microtubules and appear to regulate their polymerization, bundling, and/or stabilization in migrating neurons. In humans, mutations in DCX cause an X-linked type of lissen-cephaly known as double cortex syndrome (also called subcortical band heterotopia), whereas mutations in the Lisl gene cause classic lissencephaly, the Miller-Dieker syndrome (Ross and Walsh, 2001).

As expected from their crucial function in nuclear movement, proteins involved in this process are highly conserved throughout evolution. For example, the Lisl homologue in the filamentous fungus Aspergillus nidulans is a nuclear migration gene. During development of the fungus, cells become multinucleated through several rounds of divisions and it becomes crucial that nuclei disperse uniformly within the cell for normal growth to occur. This process of nuclear migration in fungi also depends on the network of microtubules and is regulated by proteins that associate with the micro-tubules, such as that encoded by the nudF gene. (Proteins related to nuclear movement in A. nidulans were isolated through a screen for nuclear distribution mutants, for which they are named.) nudF shares 42% sequence identity with Lisl and both genes are considered orthologues. Analysis of other nuclear distribution mutants similar to nudF has helped to define the molecular mechanisms mediating the function of this protein in fungi and, by extension, in migrating neurons. For example, nudF closely interacts with nudA, a gene that encodes the heavy chain of cytoplasmic dynein and is directly involved in nuclear translocation. Another protein that appears to act as a downstream effector of nudF is NUDE, two homologues of which have been isolated in mammals, mNudE and NUDEL. Both of these proteins localize to the MTOC and appear to be important in controlling the movement of the nucleus through their association with other proteins, such as 7-tubulin or dynein. Indeed, Lis1, dynein, or NUDEL loss of function results in defects of centrosome-nucleus coupling during neuronal migration (Shu et al., 2004; Tanaka et al., 2004). In summary, these findings illustrate how the identification of homologous proteins in model systems such as A. nidulans is greatly contributing to the identification of the function of vertebrate proteins associated with neuronal migration and, more specifically, nucleokinesis (Feng and Walsh, 2001).

In addition to proteins that directly associate with the microtubule network encaging the nucleus during nuclear translocation, other signaling proteins appear to be crucial for normal nucleokinesis. One of these proteins is Cdk5, a serine/threonine kinase that phosphorylates proteins that maintain cytoske-letal structures and promote cell motility. Mice deficient in Cdk5 or its activating subunits, p35 and p39, exhibit prominent laminar defects in the cerebral cortex, suggesting that this signaling pathway is crucial for neuronal migration (reviewed in Dhavan and Tsai, 2001). For instance, NUDEL is a physiological substrate of Cdk5 (Niethammer et al.,

2000; Sasaki et al., 2000). Another case is the focal adhesion kinase (FAK), which is localized in a Cdk5 phosphorylation-dependent manner to the perinuclear network of microtubules where it contributes to normal nuclear movement (Xie et al., 2003). Another example is mPar6a, a protein that associates with different forms of protein kinase C and localizes to the MTOC, where it contributes to promote the polarization of the centrosome in the direction of the movement. Because movement of the centrosome precedes that of the nucleus itself, the function of proteins such as mPar6a is essential for determining the direction of nucleokinesis.

In summary, multiple components of the cellular machinery involved in nucleokinesis have been already identified and a model for understanding nucleokinesis in migrating neurons is starting to emerge (Figure 1). As in the past few years, it is expected that the discovery of other proteins involved in this process may arise through additional homology analyses, since it is clear now that the cellular mechanisms underlying nuclear migration are similar throughout evolution, from unicellular organisms to humans.

Was this article helpful?

0 0

Post a comment