Receptor tyrosine kinases (RTKs) play important roles in cellular proliferation, differentiation, and survival. Reverse transcriptase-polymerase chain reactions (RT-PCR) from enriched embryonic day 5 (E5) chick motoneurons were performed by panning to identify RTKs involved in the early development of motoneuron. Cek8, a member of the eph family, is specifically expressed on motoneurons at the brachial and lumbar segments of the spinal cord that innervate limb muscles; Cek8 disappears after the naturally occurring cell death period (E6-E11). Immunohistochemistry using an anti-Cek8 monoclonal antibody showed the localization of Cek8 protein at the cell bodies and axonal fibers of motoneurons and muscles. The unique expression of Cek8 suggests its involvement in cellular survival or cell-cell interactions for specific subpopulations of developing motoneurons (Ohta, 1996).
Many Eph-related receptor tyrosine kinases, and their numerous membrane-bound ligands, can all be grouped into only two major specificity subclasses. Receptors in a given subclass bind most members of a corresponding ligand subclass. The physiological relevance of these groupings is suggested by viewing the collective distributions of all members of a subclass. These composite distributions, in contrast with less informative patterns seen with individual members of the family, reveal that the developing embryo is subdivided into domains defined by reciprocal and apparently mutually exclusive expression of a receptor subclass and its corresponding ligands. Receptors seem to encounter their ligands only at the interface between these domains. This reciprocal compartmentalization implicates the Eph family in the formation of spatial boundaries that may help to organize the developing body plan (Gale, 1996).
Recent studies have implicated Eph-related receptor tyrosine kinases and their membrane-bound ligands in restricting or stimulating the movement of cells and axons. Members of these large families of receptors and ligands fall into two major binding specificity classes, in which the GPI-anchored subgroup of ligands can each bind to all members of a subgroup of receptors, whereas the transmembrane ligands interact with a distinct subgroup of receptors. Analysis of expression patterns is therefore important in order to understand which receptor-ligand interactions occur in vivo. Mouse orthologs of five members of the ligand family have been cloned and their developmental expression has been analysed in detail, both in comparison with one other, and with the receptor specificity class with which they interact. B61, AL-1/RAGS, LERK4, and ELF-1, members of the GPI-anchored subgroup of ligands, have both distinct and overlapping aspects to their expression in early mesoderm, somites, and branchial arches; in complex, dynamic patterns in the limb, and in spatial domains and specific neurons in the CNS. Similarly, Elk-L is expressed in hindbrain segments, the roof plate, and floor plate, overlapping in these locations the expression of other transmembrane ligands, but Elk-L has distinct expression in somites. The expression domains of ligands are complementary to those of the corresponding receptors in a number of tissues, including the midbrain, hindbrain, and differentiating limbs, consistent with potential roles in restricting cell movement. There are some overlaps in the expression of receptors and ligands, for example, in somites and the early limb. Taken together with previous studies showing that Eph-related receptors also have distinct but overlapping expression patterns, these data indicate that each ligand may have stage- and tissue-specific interactions with an individual member or multiple members of the receptor family (Flenniken, 1996).
Neuron-target interaction is a key feature in the establishment of neuronal networks. However, the underlying mechanism remains unclear. At the time of target innervation, Bsk, an eph family receptor, is expressed at high levels in several brain regions, including the hippocampus, olfactory bulb, and retina. To study whether the ligands are expressed in the target tissues, the expression of Bsk ligands was examined using a ligand-affinity probe, Bsk-AP, which consisted of the extracellular domain of Bsk fused in frame with a human placental alkaline phosphatase. These analyses show that the ligands are expressed at high levels in the developing septum, hypothalamus, olfactory neural epithelium, and tectum. In situ hybridization studies reveal that at least three different factors are responsible for the Bsk-AP binding. In the septum, Elf-1, Lerk3 (Eff-2), and AL-1/Lerk7 are transcribed. In the hypothalamus, AL-1/Lerk7 is the ligand detected by Bsk-AP. In the olfactory system, high levels of Lerk3 are detected in the sensory neurons. Both Elf-1 and AL-1/Lerk7 are present in the tectum. These ligand-positive areas are known to be anatomically connected to Bsk-expressing regions. These observations strongly suggest that Bsk and the ligands participate in neuron-target interactions in multiple systems and provide support for their involvement in topographic projection (Zhang, 1996).
The task of organizing the vast array of longitudinal axons that constitute the ventral and lateral funiculi, which project alongside or near the vertebrate midline of the developing vertebrate spinal cord, is likely to be relegated to a number of discrete repellent guidance cues that act upon axons emanating from multiple populations of functionally distinct interneurons. The Eph family of receptor tyrosine kinases are particularly good candidates since they interact exclusively with membrane-associated ligands, exhibit dynamic and spatially restricted expression patterns in the developing central nervous system, and together with the ephrins, mediate axonal patterning and pathway selection through contact-dependent repulsion in a variety of neural systems. Given their transmembrane structure, B-class ephrins are especially suited to function as highly localized, contact-dependent repellents. Although previous studies have documented a role for these proteins in repulsive axon guidance, the function of the transmembrane ephrins in the spinal cord proper are presently unknown. mRNA encoding all three transmembrane ligands is expressed in the floor plate during commissural axon pathfinding. Furthermore, B-class ephrin protein is tightly localized to the lateral floor plate margins, in immediate proximity to longitudinal fiber tracts formed by decussated commissural axons. Strikingly, expression of B-class Eph receptors is detected on only those segments of commissural axons that have crossed the floor plate and turned into the longitudinal axis. It is proposed that the onset of EphB receptor expression subsequent to midline crossing plays an important role in maintaining the longitudinal trajectory of commissural axons through a repulsive interaction with transmembrane ephrins situated at the lateral floor plate boundaries. In support of this interpretation, it is shown that all three B-class ephrins can induce the collapse of a subset of commissural growth cones in vitro. This is the first identification of a membrane-associated factor that directly promotes the collapse of vertebrate commissural growth cones (Imondi, 2000).
To identify molecules involved in neurogenesis, monoclonal antibodies were raised against embryonic day 12.5 mouse telencephalon. One antibody, monoclonal antibody 25H11, stains predominantly the ventricular zone of the anterior and lateral telencephalon. Purification of the 25H11 antigen, a 47 kDa integral membrane protein, from mouse telencephali reveals its identity with ephrin B1. Ephrin B1 appears at the onset of neocortical neurogenesis, being first expressed in neuron-generating neuroepithelial cells and rapidly thereafter in virtually all neuroepithelial cells. Expression of ephrin B1 persists through the period of neocortical neurogenesis and is downregulated thereafter. Ephrin B1 is present on the ventricular as well as basolateral plasma membrane of neuroepithelial cells and exhibits a ventricular-high to pial-low gradient across the ventricular zone. Expression of ephrin B1 is also detected on radial glial cells, extending all the way to their pial endfeet, and on neurons in the mantle/intermediate zone but not in the cortical plate. These results suggest that ephrin B1, presumably via ephrin-Eph receptor signaling, has a role in neurogenesis. Given the ventricular-to-pial gradient of ephrin B1 on the neuroepithelial cell surface and its known role in cell migration in other systems mediated by its repulsive properties, it is proposed that ephrin B1 may be involved in the migration of newborn neurons out from the ventricular zone toward the neocortex (Stuckmann, 2001).
Development of the tectum and the cerebellum is induced by a reciprocal inductive signaling between their respective primordia, the midbrain and the midbrain/hindbrain boundary (MHB). It is of interest to identify molecules that function in and downstream of this reciprocal signaling. Overexpression of LIM domain of the transcription factor Islet-3 (LIM Isl-3) leads to inhibition of this reciprocal signaling and to resultant defects in tectal and cerebellar development. Genes were sought that may be either up- or down-regulated by overexpression of LIMIsl-3 by comparing the gene expression profiles in the midbrain and the MHB of normal embryos and embryos in which Islet-3 function is repressed, using a combination of ordered differential display and whole-mount in situ hybridization. Among genes identified in this search, two cDNA fragments encode Wnt1 and FGF8, which are already known to be essential for the reciprocal signaling between the midbrain and the MHB, confirming the effectiveness of this strategy. Four other partial cDNA clones were identified that were specifically expressed around the MHB, ten cDNAs specifically expressed in the tectum, and three cDNAs expressed in neural crest cells, including those derived from the midbrain level. The ephrin-A3 gene is specifically expressed in posterior tectum in a gradient that decreases anteriorly. Although ephrin-A2 and ephrin-A5 have been reported to be expressed in the corresponding region in mouse embryos (the superior/inferior colliculi), mouse ephrin-A3 is not expressed prominently in this region, suggesting that the role of ephrin-A3 in brain development may have been altered in the process of brain evolution (Hirate, 2001).
Graded expression of the Eph receptor EphA3 in the retina and its two ligands, ephrin A2 and ephrin A5 in the optic tectum, the primary target of retinal axons, has been implicated in the formation of the retinotectal projection map. Two homeobox containing genes, SOHo1 and GH6, are expressed in a nasal-high, temporal-low pattern during early retinal development, and thus in opposing gradients to EphA3. SOHo1 and GH6 belong to the HMX family of related homeobox genes (Drosophila homolog: H6-like-homeobox). Hmx genes share a conserved homeobox and have been identified in a number of species. Hmx genes are predominantly, but not exclusively, expressed in sensory organs, branchial arches and the rostral central nervous system (CNS). Hmx3/Nkx5.1 is critical for inner ear development. Retroviral misexpression of SOHo1 or GH6 completely and specifically represses EphA3 expression in the neural retina, but not in other parts of the central nervous system, such as the optic tectum. Under these conditions, some temporal ganglion cell axons overshoot their expected termination zones in the rostral optic tectum, terminating aberrantly at more posterior locations. However, the majority of ganglion cell axons map to the appropriate rostrocaudal locations, although they form somewhat more diffuse termination zones. These findings indicate that other mechanisms, in addition to differential EphA3 expression in the neural retina, are required for retinal ganglion axons to map to the appropriate rostrocaudal locations in the optic tectum. They further suggest that the control of topographic specificity along the retinal nasal-temporal axis is already split into several independent pathways at a very early time in development (Schulte, 2000).
Mesenchymal patterning is an active process whereby genetic commands coordinate cell adhesion, sorting and condensation, and thereby direct the formation of morphological structures. In mice that lack the Hoxa13 gene, the mesenchymal condensations that form the autopod skeletal elements are poorly resolved, resulting in missing digit, carpal and tarsal elements. In addition, mesenchymal and endothelial cell layers of the umbilical arteries (UAs) are disorganized, resulting in their stenosis and in embryonic death. To further investigate the role of Hoxa13 in these phenotypes, a loss-of-function allele was generated in which the GFP gene was targeted into the Hoxa13 locus. This allele allows FACS isolation of mesenchymal cells from Hoxa13 heterozygous and mutant homozygous limb buds. Hoxa13GFP expressing mesenchymal cells from Hoxa13 mutant homozygous embryos are defective in forming chondrogenic condensations in vitro. Analysis of pro-adhesion molecules in the autopod of Hoxa13 mutants reveals a marked reduction in EphA7 expression in affected digits, as well as in micromass cell cultures prepared from mutant mesenchymal cells. Finally, antibody blocking of the EphA7 extracellular domain severely inhibits the capacity of Hoxa13GFP heterozygous cells to condense and form chondrogenic nodules in vitro, which is consistent with the hypothesis that reduction in EphA7 expression affects the capacity of Hoxa13–/– mesenchymal cells to form chondrogenic condensations in vivo and in vitro. EphA7 and EphA4 expression are also decreased in the mesenchymal and endothelial cells that form the umbilical arteries in Hoxa13 mutant homozygous embryos. These results suggest that an important role for Hoxa13 during limb and UA development is to regulate genes whose products are required for mesenchymal cell adhesion, sorting and boundary formation (Stadler, 2001).
Chick brain factor 1 (CBF1), a nasal retina-specific winged-helix transcription factor, is known to prescribe the nasal specificity that leads to the formation of the precise retinotectal map, especially along the anteroposterior (AP) axis. However, its downstream topographic genes and the molecular mechanisms by which CBF1 controls the expression of them have not been elucidated. Misexpression of CBF1 represses the expression of EphA3 and CBF2, and induces that of SOHo1, GH6, ephrin A2 and ephrin A5. CBF1 controls ephrin A5 by a DNA binding-dependent mechanism, ephrin A2 by a DNA binding-independent mechanism, and CBF2, SOHo1, GH6 and EphA3 by dual mechanisms. BMP2 expression begins double-gradiently (varying in both naso-temporal and ventral-dorsal axes) in the retina from E5 in a complementary pattern to Ventroptin expression. Ventroptin antagonizes BMP2 as well as BMP4. CBF1 interferes in BMP2 signaling and thereby induces expression of ephrin A2. These data suggest that CBF1 is located at the top of the gene cascade for the regional specification along the nasotemporal (NT) axis in the retina and distinct BMP signals play pivotal roles in the topographic projection along both axes (Takahashi, 2003).
As axons grow past intermediate targets, they change their responsiveness to guidance cues. Local upregulation of receptor expression is involved, but the mechanisms for this are not clear. Protein synthesis can be traced within individual axons by introducing RNAs encoding visualizable reporters. Individual severed axons and growth cones can translate proteins and also export them to the cell surface. As axons reach the spinal cord midline, EphA2 is among the receptors upregulated on at least some distal axon segments. Midline reporter upregulation is recapitulated by part of the EphA2 mRNA 3' untranslated region, which is highly conserved and includes known translational control sequences. These results show axons contain all the machinery for protein translation and cell surface expression, and they reveal a potentially general and flexible RNA-based mechanism for regulation localized within a subregion of the axon (Brittis, 2002).
Sek4 and Nuk are members of the Eph-related family of receptor protein-tyrosine kinases. These receptors interact with a set of cell surface ligands that have recently been implicated in axon guidance and fasciculation. The formation of the corpus callosum and anterior commissure, two major commissural axon tracts that connect the two cerebral hemispheres, is critically dependent on Sek4 and Nuk. While mice deficient in Nuk exhibit defects in pathfinding of the anterior commissure axons, sek4 mutants have defects in corpus callosum formation. The phenotype in both axon tracts is markedly more severe in sek4/nuk1 double mutants, indicating that the two receptors act in a partially redundant fashion. sek4/nuk1 double mutants also exhibit specific guidance and fasciculation defects of diencephalic axon tracts. Moreover, while mice singly deficient in either Sek4 or Nuk are viable, most sek4/nuk1 double mutants die immediately after birth primarily due to a cleft palate. These results demonstrate essential and cooperative functions for Sek4 and Nuk in establishing axon pathways in the developing brain, and during the development of facial structures (Orioli, 1996).
Tyro4 is a member of the eph family of receptor protein-tyrosine kinases. Tyro4 is the rat homolog of mouse Mek4 and chick Cek4. An evolutionarily conserved pattern of expression for Tyro4, Mek4, and Cek4 has been demonstrated. Most strikingly, this receptor is specifically expressed, in all three species, in a subset of motor neurons in the medial motor column and in a subset of axial, but not limb, muscles. Mek4 has previously been ascribed a role in guiding retinal axons to their targets in the optic tectum. These results extend the purported role of Mek4 in axon guidance to include motor neurons of the medial motor column (Kilpatrick, 1996).
Mice that are homozygous for a mutation that disrupts the gene encoding EphA8, a member of the Eph family of tyrosine protein kinase receptors, previously known as Eek. These mice develop to term, are fertile and do not display obvious anatomical or physiological defects. The mouse ephA8/eek gene is expressed primarily in a rostral to caudal gradient in the developing tectum. Axonal tracing experiments have revealed that in these mutant mice, axons from a subpopulation of tectal neurons located in the superficial layers of the superior colliculus do not reach targets located in the contralateral inferior colliculus. Moreover, ephA8/eek null animals display an aberrant ipsilateral axonal tract that projects to the ventral region of the cervical spinal cord. Retrograde labeling reveals that these abnormal projections originate from a small subpopulation of superior colliculus neurons that normally express the ephA8/eek gene. These results suggest that EphA8/Eek receptors play a role in axonal pathfinding during development of the mammalian nervous system (Park, 1997).
AL-1 is a glycosylphosphatidylinositol-linked ligand for the Eph-related receptor, REK7. It shows that a REK7-IgG fusion protein can block axon bundling in co-cultures of cortical neurons on astrocytes, suggesting a role for REK7 and AL-1 in axon fasciculation. Subsequent identification of RAGS, the chick homolog of AL-1, as a repellent axon guidance molecule in the developing chick visual system has led to speculation that AL-1, expressed on astrocytes, provides a repellent stimulus for cortical axons, inducing them to bundle as an avoidance mechanism. Using a growth cone collapse assay to test this hypothesis, it has been shown that a soluble AL-1-IgG fusion protein is a potent collapsing factor for embryonic rat cortical neurons. The response is strongly correlated with REK7 expression, implicating REK7 as a receptor mediating AL-1-induced collapse. Morphological collapse is preceded by an AL-1-IgG-induced reorganization of the actin cytoskeleton that resembles the effects of cytochalasin D. This suggests a pathway whereby REK7 activation by AL-1 leads to perturbation of the actin cytoskeleton, possibly by an effect on actin polymerization, followed by growth cone collapse. AL-1-IgG causes collapse of rat hippocampal neurons and rat retinal ganglion cells. These data suggest a role for REK7 and AL-1 in the patterning of axonal connections in the developing cortex, hippocampus and visual system (Meima, 1997a).
The transmembrane (TM) subfamily of Eph ligands and their receptors have been implicated in axon pathfinding and in pattern formation during embryogenesis. These functions are thought to involve repulsive interactions but this has not been demonstrated directly. A growth cone collapse assay has been used to determine if the TM ligands Lerk2 and HtkL have repellant guidance activity. Lerk2, but not HtkL, is a collapsing factor for a subset of embryonic cortical neurons. Analysis of the effects of Lerk2 on both the morphology and the cytoskeleton of cortical neurons suggests a mechanism of action different from that of AL-1, a GPI-linked Eph ligand having similar repellant activity. Treatment with Lerk2 disrupts the organization of both the actin cytoskeleton and the microtubules and induces the formation of swellings in the center of the growth cone and along the axon. Measurement of the relative F-actin concentrations in the neurites and soma indicate that F-actin levels in the neurites decrease while those in the soma increase, with the net F-actin content of the neuron remaining unchanged. In contrast, prolonged treatment with AL-1 leads to a net loss of F-actin, consistent with the hypothesis that AL-1 acts by perturbing actin polymerization. These results provide evidence that the ectodomain of Lerk2 functions as a repellant guidance cue and show that, despite overlapping specificities in vitro, the biological activities of related ligands are not necessarily overlapping. Further, TM and GPI-linked Eph ligands appear to exert repellant activity by different mechanisms, opening up the possibility that they may have different effects on growth cones in vivo (Meima, 1997b).
During the past few years, evidence has accumulated that activation of EphA kinases on the axons of retinal ganglion cells by ephrins-A on cells of the optic tectum (superior colliculus) plays a critical role in mapping the rostrocaudal axis of the retina onto the corresponding axis of the tectum. The similarities between topographic maps in the motor and visual systems has suggested that similar molecular mechanisms might underlie them. Motor axons form topographic maps on muscles: rostral motor pools innervate rostral muscles, and rostral portions of motor pools innervate rostral fibers within their targets. Subfamily A ephrins are implicated in this topographic mapping because: (1) developing muscles express all five of the ephrin-A genes; (2) rostrally and caudally derived motor axons differ in sensitivity to outgrowth inhibition by ephrin-A5; (3) the topographic map of motor axons on the gluteus muscle is degraded in transgenic mice that overexpress ephrin-A5 in muscles; (4) topographic mapping is impaired in muscles of mutant mice lacking ephrin-A2 plus ephrin-A5. Thus, ephrins mediate or modulate positionally selective synapse formation. In addition, the rostrocaudal position of at least one motor pool is altered in ephrin-A5 mutant mice, indicating that ephrins affect nerve-muscle matching by intraspinal as well as intramuscular mechanisms. If the conclusion that ephrins are directly involved in nerve-muscle matching is accepted, new questions arise about how they act. These can be divided into issues of mapping and synaptogenesis. In terms of mapping, one plausible model is that muscle fibers express different levels or combinations of ephrins in accordance with their positions and that motor axons express different levels or combinations of Eph kinases (the likely ephrin receptors) in accordance with their positions. For example, levels of expression might be graded along the rostrocaudal axis, as appears to be the case in the retinotectal system. Alternatively, patterns of ephrin and Eph kinase expression might describe a more complex combinatorial code, as suggested by the discrete identities of motor pools and muscles. Indeed, whereas Eph kinase and ephrin levels are graded along the rostrocaudal axis in the visual system, there is currently no evidence for graded expression of ligands or receptors in the motor system. Another possiblity is that interactions between ephrins and their receptors are permissive rather than instructive, enabling positional information to be transmitted by other molecules. It is important to emphasize that the data do not distinguish permissive versus instructive roles for ephrins (Feng, 2000).
In the embryonic visual system, EphA receptors are expressed on both temporal and nasal retinal ganglion cell axons. Only the temporal axons, however, are sensitive to the low concentrations of ephrin-A ligands found in the anterior optic tectum. The poor responsiveness of nasal axons to ephrin-A ligands, which allows them to traverse the anterior tectum and reach their targets in the posterior tectum, has been attributed to constitutive activation of the EphA4 receptor expressed in these axons. EphA4 is highly expressed throughout the retina, but is preferentially phosphorylated on tyrosine (activated) in nasal retina. In a screen for EphA4 ligands expressed in chicken embryonic retina, a novel ephrin, ephrin-A6, has been identified. Like ephrin-A5, ephrin-A6 has high affinity for EphA4 and activates this receptor in cultured retinal cells. In the embryonic day 8 (E8) chicken visual system, ephrin-A6 is predominantly expressed in the nasal retina and ephrin-A5 in the posterior tectum. Thus, ephrin-A6 has the properties of a ligand that activates the EphA4 receptor in nasal retinal cells. Ephrin-A6 binds with high affinity to several other EphA receptors as well and causes growth cone collapse in retinal explants, demonstrating that it can elicit biological responses in retinal neurons. Ephrin-A6 expression is high at E6 and E8, when retinal axons grow to their tectal targets, and gradually declines at later developmental stages. The asymmetric distribution of ephrin-A6 in retinal cells, and the time course of its expression, suggest that this new ephrin plays a role in the establishment of visual system topography (Menzel, 2001).
The role of the Eph family of receptor tyrosine kinases and their ligands in the establishment of the vomeronasal projection in the mouse has been investigated. The data show intriguing differential expression patterns of ephrin-A5 on vomeronasal axons and of EphA6 in the accessory olfactory bulb (AOB), such that axons with high ligand concentration project onto regions of the AOB with high receptor concentration and vice versa. These data suggest a mechanism for development of this projection that is the opposite of the repellent interaction between Eph receptors and ligands observed in other systems. In support of this idea, when given the choice of whether to grow on lanes containing EphA-Fc/laminin or Fc/laminin protein (in the stripe assay), vomeronasal axons prefer to grow on EphA-Fc/ laminin. Analysis of ephrin-A5 mutant mice reveal a disturbance of the topographic targeting of vomeronasal axons to the AOB. In summary, these data, which are derived from in vitro and in vivo experiments, indicate an important role of the EphA family in setting up the vomeronasal projection (Knoll, 2001a).
Growing axons follow highly stereotypical pathways, guided by a variety of attractive and repulsive cues, before establishing specific connections with distant targets. A particularly well-known example that illustrates the complexity of axonal migration pathways involves the axonal projections of motor neurons located in the motor cortex. These projections take a complex route during which they first cross the midline, then form the corticospinal tract, and ultimately connect with motor neurons in the contralateral side of the spinal cord. These obligatory contralateral connections account for why one side of the brain controls movement on the opposing side of the body. The netrins and slits provide well-known midline signals that regulate axonal crossings at the midline. A member of the ephrin family, ephrin-B3, also plays a key role at the midline to regulate axonal crossing. In particular, ephrin-B3 acts as the midline barrier that prevents corticospinal tract projections from recrossing when they enter the spinal gray matter. In ephrin-B3-/- mice, corticospinal tract projections freely recross in the spinal gray matter, such that the motor cortex on one side of the brain now provides bilateral input to the spinal cord. This neuroanatomical abnormality in ephrin-B3-/- mice correlates with loss of unilateral motor control, yielding mice that simultaneously move their right and left limbs and thus have a peculiar hopping gait quite unlike the alternate step gait displayed by normal mice. The corticospinal and walking defects in ephrin-B3-/- mice resemble those recently reported for mice lacking the EphA4 receptor, which binds ephrin-B3 as well as other ephrins, suggesting that the binding of EphA4-bearing axonal processes to ephrin-B3 at the midline provides the repulsive signal that prevents corticospinal tract projections from recrossing the midline in the developing spinal cord (Kullander, 2001a).
To investigate Eph-ephrin bidirectional signaling, a series of mutations were generated in the ephrin-B3 locus. The absence of both forward and reverse signaling results in mice with mirror movements as typified by a hopping locomotion. The corticospinal tract is defective since axons fail to respect the midline bound high-fidelity connection between the motor cortex and contralateral and ipsilateral motor neuron populations. A second mutation that expresses a truncated ephrin-pathfinding protein lacking Ephrin-B3's cytoplasmic domain does not lead to hopping, indicating that reverse signaling is not required for corticospinal innervation. Ephrin-B3 is concentrated at the spinal cord midline, while one of its receptors, EphA4, is expressed in postnatal cortico-spinal neurons as their fibers pathfind down the contralateral spinal cord. These data indicate ephrin-B3 functions as a midline-anchored repellent to stimulate forward signaling in EphA4-expressing axons (Yokoyama, 2001).
In both invertebrate and lower vertebrate species, decussated commissural axons travel away from the midline and assume positions within distinct longitudinal tracts. In the developing chick and mouse spinal cord, most dorsally situated commissural neuron populations extend axons across the ventral midline and through the ventral white matter along an arcuate trajectory on the contralateral side of the floor plate. Within the dorsal (chick) and intermediate (mouse) marginal zone, commissural axons turn at a conserved boundary of transmembrane ephrin expression, adjacent to which they form a discrete ascending fiber tract. In vitro perturbation of endogenous EphB-ephrinB interactions results in the failure of commissural axons to turn at the appropriate dorsoventral position on the contralateral side of the spinal cord; consequently, axons inappropriately invade more dorsal regions of B-class ephrin expression in the dorsal spinal cord. Taken together, these observations suggest that B-class ephrins act locally during a late phase of commissural axon pathfinding to specify the dorsoventral position at which decussated commissural axons turn into the longitudinal axis (Imondi, 2001).
The development of connections between thalamic afferents and their cortical target cells occurs in a highly precise manner. Thalamic axons enter the cortex through deep cortical layers, then stop their growth in layer 4 and elaborate terminal arbors specifically within this layer. The mechanisms that underlie target layer recognition for thalamocortical projections are not known. The growth patterns have been compared of thalamic explants cultured on membrane substrates purified from cortical layer 4, the main recipient layer for thalamic axons, and cortical layer 5, a non-target layer. Thalamic axons exhibit a reduced growth rate and an increased branching density on their appropriate target membranes compared with non-target substrate. When confronted with alternating stripes of both membrane substrates, thalamic axons grow preferentially on their target membrane stripes. Enzymatic treatment of cortical membranes has revealed that growth, branching and guidance of thalamic axons are independently regulated by attractive and repulsive cues differentially expressed in distinct cortical layers. These results indicate that multiple membrane-associated molecules collectively contribute to the laminar targeting of thalamic afferents. Furthermore, it was found that interfering with the function of Eph tyrosine kinase receptors and their ligands (ephrins) abolishes the preferential branching of thalamic axons on their target membranes, and that recombinant ephrin-A5 ligand elicites a branch-promoting activity on thalamic axons. It is concluded that interactions between Eph receptors and ephrins mediate branch formation of thalamic axons and thereby may play a role in the establishment of layer-specific thalamocortical connections (Mann, 2002).
Motor neurons in the ventral neural tube project axons specifically to their target muscles in the periphery. Although many of the transcription factors that specify motor neuron cell fates have been characterized, less is understood about the mechanisms that guide motor axons to their correct targets. Ectopic expression of EphA4 receptor tyrosine kinase alters the trajectories of a specific population of motor axons in the avian hindlimb. Most motor neurons in the medial portion of the lateral motor column (LMC) extend their axons aberrantly in the dorsal nerve trunk at the level of the crural plexus, in the presence of ectopic EphA4. This misrouting of motor axons is not accompanied by alterations in motor neuron identity, settling patterns in the neural tube, or the fasciculation of spinal nerves. However, ectopic EphA4 axons do make errors in pathway selection during sorting in the plexus at the base of the hindlimb. These results suggest that EphA4 in motor neurons acts as a population-specific guidance cue to control the dorsal trajectory of their axons in the hindlimb (Eberhart, 2002).
Neural maps in the vertebrate central nervous system often show discontinuously segregated, domain-to-domain patterns. However, the molecular mechanism that establishes such maps is not well understood. In the chicken olivocerebellar system, EphA receptors and ephrin-As are expressed with distinct levels and combinations in mapping domains. When ephrin-A2 is retrovirally overexpressed in the cerebellum, the olivocerebellar map is disrupted, excluding axons with high receptor activity from ectopic expression domains. Conversely, overexpression of a truncated EphA3 receptor in the cerebellum reduces endogenous ligand activity to undetectable levels and causes aberrant mapping, with high receptor axons invading high ligand domains. In vitro, ephrin-A2 inhibits outgrowth of inferior olive axons in a region-specific manner. These results suggest that Eph receptors and ephrins constitute domain-specific positional information, and the spatially accurate receptor-ligand interaction is essential to guide inferior olive axons to their correct target domains (Nishida, 2002).
The mechanisms generating precise connections between specific thalamic nuclei and cortical areas remain poorly understood. Using axon tracing analysis of ephrin/Eph mutant mice, in vivo evidence is provided that Eph receptors in the thalamus and ephrins in the cortex control intra-areal topographic mapping of thalamocortical (TC) axons. In addition, the same ephrin/Eph genes unexpectedly control the inter-areal specificity of TC projections through the early topographic sorting of TC axons in an intermediate target, the ventral telencephalon. These results constitute the first identification of guidance cues involved in inter-areal specificity of TC projections and demonstrate that the same set of mapping labels is used differentially for the generation of topographic specificity of TC projections between and within individual cortical areas (Dufour, 2003).
Molecular mechanisms generating the topographic organization of corticothalamic (CT) circuits, which comprise more than three-quarters of the synaptic inputs onto sensory relay neurons, and their interdependence with thalamocortical (TC) axon development, are unknown. Using in utero electroporation-mediated gene transfer, EphA7-mediated signaling on neocortical axons was shown to control the within-nucleus topography of CT projections in the thalamus. Notably, CT axons that misexpress EphA7 do not shift the relative positioning of their pathway within the subcortical telencephalon (ST), indicating that they do not depend upon EphA7/ephrin-A signaling in the ST for establishing this topography. Moreover, misexpression of cortical EphA7 results in disrupted topography of CT projections, but unchanged inter- and intra-areal topography of TC projections. These results support a model in which EphA/ephrin-A signaling controls independently the precision with which CT and TC projections develop, yet is essential for establishing their topographic reciprocity (Torii, 2005).
How can TC and CT projections achieve their reciprocal organization without interdependent controls? Based on the present results and studies suggesting that the intra-areal topography of TC projections is regulated by an EphA gradient in thalamic nuclei and an ephrin-A gradient in target cortical areas, a model is proposed in which complementary gradients of EphA and ephrin-A activity, located between as well as within specific cortical areas and their corresponding thalamic nuclei, produce reciprocal signaling between these structures. In this model, however, the EphA/ephrin-A signaling controls the topography of TC and CT projections independently, yet establishes essential functional topographic reciprocity (Torii, 2005).
In this model, functional gradients of EphA receptors and ephrin-A ligands are required to be complementary both between and within each reciprocal target. Specific expression patterns of EphAs and ephrin-As appear before CT and TC axons reach their targets. Together with recent evidence for intrinsic regulation of EphA7 and ephrin-A5 expression within the neocortex by transcription factors or morphogens, gradients of EphAs and ephrin-As are likely established by intrinsic regulation within the neocortex and DT independently. Complementary patterns might be achieved by reciprocal transcriptional regulation for each by the same transcription factor. For example, in dorsoventral patterning of the retina, Tbx5 enhances transcription of ephrin-Bs and represses EphBs. In contrast, Vax2 enhances the transcription of EphBs and represses ephrin-Bs. In addition to transcriptional regulation, repressive modulation of EphA function by coexpressed ephrin-A, as shown on retinal ganglion cell axons, might sharpen their functional complementarity where their expression overlaps. Reciprocal and topographic neural connections are found in many other systems throughout the brain. It will be important to determine whether this independent, but reciprocal, EphA/ephrin-A signaling model is applicable generally to other systems (Torii, 2005).
Ephs regulate growth cone repulsion, a process controlled by the actin cytoskeleton. The guanine nucleotide exchange factor (GEF) ephexin1 (Drosophila homolog: CG3799) interacts with EphA4 and has been suggested to mediate the effect of EphA on the activity of Rho GTPases, key regulators of the cytoskeleton and axon guidance. Using cultured ephexin1-/- mouse neurons and RNA interference in the chick, it was found that ephexin1 is required for normal axon outgrowth and ephrin-dependent axon repulsion. Ephexin1 becomes tyrosine phosphorylated in response to EphA signaling in neurons, and this phosphorylation event is required for growth cone collapse. Tyrosine phosphorylation of ephexin1 enhances ephexin1ís GEF activity toward RhoA while not altering its activity toward Rac1 or Cdc42, thus changing the balance of GTPase activities. These findings reveal that ephexin1 plays a role in axon guidance and is regulated by a switch mechanism that is specifically tailored to control Eph-mediated growth cone collapse (Sahin, 2005).
Ephrin-As and their receptors, EphAs, are expressed in the developing cortex where they may act to organize thalamic inputs. This study mapped the visual cortex (V1) in mice deficient for ephrin-A2, -A3, and -A5 functionally [ephrin-A triple knock out (TKO)], using intrinsic signal optical imaging and microelectrode recording, and structurally, by anatomical tracing of thalamocortical projections. V1 is shifted medially, rotated, and compressed and its internal organization is degraded. Expressing ephrin-A5 ectopically by in utero electroporation in the lateral cortex shifts the map of V1 medially, and expression within V1 disrupts its internal organization. These findings indicate that interactions between gradients of EphA/ephrin-A in the cortex guide map formation, but that factors other than redundant ephrin-As are responsible for the remnant map. Together with earlier work on the retinogeniculate map, the current findings show that the same molecular interactions may operate at successive stages of the visual pathway to organize maps (Cang, 2005).
The visual system has been a long-standing model in the study of topographic map development. EphA receptors and ephrin-A ligands have been shown to be necessary for the development of topographic maps from the retina to two of its subcortical targets, the superior colliculus (SC) in the midbrain and dorsal lateral geniculate nucleus (dLGN) in the thalamus. The dLGN then projects to the visual cortex. In this study, it was found that the topographic mapping of geniculocortical projections is disrupted both structurally and functionally in an ephrin-A TKO. The expression patterns of EphA receptors and ephrin-A ligands in the geniculocortical projection are consistent with EphA/ephrin-A interactions acting in thalamocortical (TC) projections in the same manner as that proposed for EphA/ephrin-A interactions in subcortical mapping. In the cases in which ephrin-A ligand was successfully misexpressed within the developing visual cortex, the functional topographic maps were defective, consistent with ephrin-A ligands acting as molecular labels in the cortex. Together, these experiments suggest that the same mapping mechanisms are used at subcortical and cortical stages, acting in each projection and target area to maintain the topographic order of visual information as it is transferred to the next stage (Cang, 2005).
Signaling by receptor tyrosine kinases (RTKs) is mediated by their intrinsic kinase activity. Typically, kinase-activating mutations result in ligand-independent signaling and gain-of-function phenotypes. Like other RTKs, Ephs require kinase activity to signal, but signaling by Ephs in vitro also requires clustering by their membrane bound ephrin ligands. The relative importance of Eph kinase activity and clustering for in vivo functions is unknown. Knockin mice expressing a mutant form of EphA4 (EphA4EE), whose kinase is constitutively activated in the absence of ephrinB ligands, are deficient in the development of thalamocortical projections and some aspects of central pattern generator rhythmicity. Surprisingly, other functions of EphA4 are regulated normally by EphA4EE, including midline axon guidance, hindlimb locomotion, in vitro growth cone collapse, and phosphorylation of ephexin1. These results suggest that signaling of Eph RTKs follows a multistep process of induced kinase activity and higher-order clustering that is different from RTKs responding to soluble ligands (Egea, 2005).
Among the diverse in vivo functions of EphA4, thalamocortical projections were not rescued by kinase-active EphA4EE. One of the possible explanations is that thalamocortical projections are controlled by ephrinA5, whereas midline guidance of CST and CPG axons is provided by ephrinB3. It is possible that the functional requirement for regulated kinase activity differs depending on whether the receptor is activated by ephrinA or ephrinB ligands. Also, the intrinsic response of thalamic neurons to EphA4 signaling may be different from that of corticospinal or spinal cord interneurons. The JM tyrosine residues that are mutated in ephA4EE, besides controlling the kinase activity of the receptor, have been shown to bind SH2 domain-containing effector proteins. It is conceivable that thalamic neurons require an EphA4 effector that binds to phosphorylated JM tyrosine residues (Egea, 2005).
A final possibility for the phenotype in thalamocortical projections may be the fact that they are topographic and EphA4 has to respond to a smooth gradient of ligand, whereas midline guidance is achieved by a step gradient of ligand. The ability to generate a graded response to ephrins likely requires a particularly tight regulation of Eph activation, including its phosphorylation state: in this context the dynamic range of sensitivity of EphA4 to ephrin gradients could be impaired in ephA4EE/EE mutants, hence leading to the topographic errors observed in vivo in the thalamocortical system. It would be interesting to set up in vitro assays challenging the EphA4EE receptor with either smooth or step gradients of the same ligand to investigate whether the regulation of kinase activity is more critical when the growth cone has to sense an ephrin gradient. In this respect, it is interesting to note that the physiology of the spinal central pattern generator (CPG) is not completely normal in the ephA4EE/EE mutants. In the normal spinal cord, EphA4 signaling is required for CPG neurons projecting ipsilaterally to an unknown target and responding to so far uncharacterized ephrins. A certain degree of drifting of firing patterns between L2 and L5 and aberrant synchrony at lumbar level 5 are in the ephA4EE/EE mutants. It is tempting to speculate that the EphA4-dependent ipsilateral projections are topographically organized and that they follow similar constraints as the thalamocortical projections. If this were true, then EphA4 would have two distinct functions in the spinal CPG: (1) to prevent aberrant midline crossing (rescued by EphA4EE) and (2) to provide ipsilateral topography (partially rescued by EphA4EE) (Egea, 2005).
The ephrin/Eph system plays a central role in neuronal circuit formation; however, its downstream effectors are poorly understood. α-chimerin Rac GTPase-activating protein mediates ephrinB3/EphA4 forward signaling. A spontaneous mouse mutation, miffy (mfy), was discovered that results in a rabbit-like hopping gait, impaired corticospinal axon guidance, and abnormal spinal central pattern generators. Using positional cloning, transgene rescue, and gene targeting, loss of α-chimerin was found to lead to mfy phenotypes similar to those of EphA4−/− and ephrinB3−/− mice. α-chimerin interacts with EphA4 and, in response to ephrinB3/EphA4 signaling, inactivates Rac, which is a positive regulator of process outgrowth. Moreover, downregulation of α-chimerin suppresses ephrinB3-induced growth cone collapse in cultured neurons. These findings indicate that ephrinB3/EphA4 signaling prevents growth cone extension in motor circuit formation via α-chimerin-induced inactivation of Rac. They also highlight the role of a Rho family GTPase-activating protein as a key mediator of ephrin/Eph signaling (Iwasato, 2007).
The expression of Cek5, a receptor-type tyrosine kinase of the Eph subclass, and its variant form Cek5+ were examined in the chick neural retina during development. Cek5 is present at high levels at all stages of retinal development examined, while Cek5+ is most abundant during differentiation. Cek5 mRNA expression and immunoreactivity are evenly distributed in the undifferentiated retina. With differentiation, Cek5 becomes concentrated in the inner and outer plexiform layers. While only moderate changes in Cek5 protein expression are observed throughout retinal development, Cek5 phosphorylation on tyrosine in vivo is dramatically increased during differentiation. This suggests that the Cek5 ligand is expressed at high levels and causes Cek5 activation. Thus, Cek5 is likely to play an active role in retinal morphogenesis, particularly during the establishment of interneuronal contacts (Pasquale, 1994).
In the retinotectal system, positional information has long been postulated to take the form of molecular gradients within both the retina and the tectum. Recent reports have implicated Mek4 in this process, as well as a member of the Eph (also named class V) family of tyrosine kinase receptors (RTKs), and two ligands: RAGS and ELF-1. QEK5, another member of the Eph family of RTKs, has been isolated from a quail cDNA library. During retinal differentiation, QEK5 transcripts accumulate in a ventral to dorsal gradient within the retinal neuroepithelium, where its expression becomes restricted to the ganglion and bipolar cell layers. Within the tectum, QEK5 transcripts are detectable in a posterior to anterior gradient in the ventricular layer and newly formed superficial layers. The pattern of QEK5 expression in the retina and tectum is distinct from that of Mek4, suggesting that complex patterns of Eph RTKs and their ligands may play a role in cell-cell interactions involved in retinotectal projections and differentiation of the central nervous system (Kenny, 1995).
Topographic maps with a defined spatial ordering of neuronal connections are a key feature of brain organization. Such maps are believed to develop in response to complementary position-specific labels in presynaptic and postsynaptic fields. However, the complementary labeling molecules are not known. In the well-studied visual map of retinal axons projecting to the tectum, the labels are hypothesized to be in gradients, without needing large numbers of cell-specific molecules. ELF-1 is a ligand for Eph family receptors. RNA hybridization shows matching expression gradients for ELF-1 in the tectum and its receptor Mek4 in the retina. Binding activity detected with alkaline phosphatase fusions of ELF-1 and Mek4 also reveals gradients and provides direct evidence for molecular complementarity of gradients in reciprocal fields. ELF-1 and Mek4 may therefore play roles in retinotectal development and have properties predicted of topographic mapping labels (Cheng, 1995).
Receptor protein tyrosine kinases of the Eph subfamily have been proposed to play roles in pattern formation based on their distribution during embryonic development. Cek5 (chicken embryo kinase 5) and Cek8 (chicken embryo kinase 8) are Eph-related kinases highly expressed in the chicken embryonic retina. To assess their potential roles in the development of the visual pathway, their distribution was examined by immunoperoxidase labeling. Cek8 is expressed throughout the pathway of the retinal ganglion cell axons, including the nerve fiber layer of the retina, optic nerve, optic chiasm, and stratum opticum of the tectum. Cek5 immunoreactivity is highly concentrated in only a portion of the optic nerve and optic chiasm; in retinal cultures, Cek5 is detected in neurons. This prompted an examination of the regional distribution of Cek5 in the developing retina and led to the observation that Cek5 is most concentrated in the ventral aspect. RT-PCR established that the differential regulation of Cek5 expression in different portions of the retina occurs at the transcriptional level. Immunoblotting analysis reveals that this unusual expression pattern is distinctive for Cek5, since three other members of the Eph subfamily, Cek4, Cek8, and Cek9, are evenly expressed across the dorsal-ventral axis of the retina. Both Cek5 and Cek8 are distributed in ways that are consistent with their regulating the outgrowth of retinal ganglion cell axons to the tectum. Furthermore, Cek5 represents the first signal transduction molecule found to exhibit the polarized pattern of expression predicted for proteins that control the specificity of the retinotectal projections (Holash, 1995).
The Eph family of receptor tyrosine kinases and their ligands can be divided into two specificity subclasses: the Eck-related receptors and their GPI-anchored ligands, and the Elk-related receptors and their transmembrane ligands. Eck- and Elk-related receptors in the retina distribute in high temporal-low nasal and high ventral-low dorsal gradients, respectively. Ligands from each subclass distribute in gradients opposing those of their corresponding receptors within the retina itself. Moreover, ligand gradients in the retina precede ganglion cell genesis. These results support an intraretinal role for Eph family members, in addition to their previously proposed role in the development of retinotectal topography. The distinct distributions of Eph family members suggest that each subclass specifies positional information along independent retinal axes (Marcus, 1996).
Visual connections to the mammalian forebrain are known to be patterned by neural activity, but it remains unknown whether the map topography of such higher sensory projections depends on axon guidance labels. Complementary expression and binding are shown for the receptor EphA5 in mouse retina and its ligands ephrin-A2 and ephrin-A5 in multiple retinal targets, including the major forebrain target, the dorsal lateral geniculate nucleus (dLGN). These ligands can act in vitro as topographically specific repellents for mammalian retinal axons and are necessary for normal dLGN mapping in vivo. The results suggest a general and economic modular mechanism for brain mapping whereby a projecting field is mapped onto multiple targets by repeated use of the same labels. They also indicate the nature of a coordinate system for the mapping of sensory connections to the forebrain (Feldheim, 1999).
The retinotectal projection map is organized in a precise retinotopic order, so that the temporo-nasal axis of the retina corresponds to the rostro-caudal axis of the tectum. en-1 and en-2, homologues of the Drosophila segment polarity gene engrailed, are expressed in a gradient along the rostro-caudal axis of the tectal anlage, and are suggested to confer caudal characteristics as the results of transplantation and ectopic engrailed (en) expression. The ligands for Eph type receptor tyrosine kinases are expressed strongly at the caudal tectum and play a role in retinotectal map formation by repulsing the temporal retinal fibers. Using the system of replication competent retroviral vector, en-2 RCAS (A/B), en-2 was misexpressed on the tectum. Elf-1 or RAGS is induced at the ectopic En-2 sites. The present results show that En-2 can regulate expression of both Elf-1 and RAGS. This suggests that the cells that express en at the early stage of tectum development acquire positional specificity as 'caudal' tectum, and these cells may later express the ligands for Eph type receptor tyrosine kinases. Therefore the temporal retinal fibers which have the receptors are repelled when they meet the ligands on the tectum (Shigetani, 1997).
In the retinotectal projection, the Eph receptor tyrosine kinase ligands ephrinA2 and ephrinA5 are differentially expressed, not only in the tectum, but also in a high-nasal-to-low-temporal pattern in the retina. Recently, retrovirally driven overexpression of ephrinA2 on retinal axons leads to topographic targeting errors of temporal axons in that they overshoot their normal termination zones in the rostral tectum and project onto the mid- and caudal-tectum. The behavior of nasal axons, however, is only marginally affected. Overexpression of ephrinA5 affects the topographic targeting behavior of both temporal and nasal axons. These data reinforce the idea that differential ligand expression on retinal axons contributes to topographic targeting in the retinotectal projection. Additionally, it has been found that ectopic expression of ephrinA2 and ephrinA5 frequently leads to pathfinding errors at the chiasm, resulting in an increased stable ipsilateral projection (Dutting, 1999).
Optic nerve formation requires precise retinal ganglion cell (RGC) axon pathfinding within the retina to the optic disc, the molecular basis of which is not well understood. At CNS targets, interactions between Eph receptor tyrosine kinases on RGC axons and ephrin ligands on target cells have been implicated in formation of topographic maps. However, studies in chick and mouse have shown that both Eph receptors and ephrins are also expressed within the retina itself, raising the possibility that this receptor-ligand family mediates aspects of retinal development. The presence of specific EphB receptors and B-ephrins in embryonic mouse retina is fully documented in this study, and evidence is provided that EphB receptors are involved in RGC axon pathfinding to the optic disc. As RGC axons begin this pathfinding process, EphB receptors are uniformly expressed along the dorsal-ventral retinal axis. This is in contrast to the previously reported high ventral-low dorsal gradient of EphB receptors later in development when RGC axons map to CNS targets. Mice lacking both EphB2 and EphB3 receptor tyrosine kinases, but not each alone, exhibit increased frequency of RGC axon guidance errors to the optic disc. In these animals, major aspects of retinal development and cellular organization appear normal, as do the expression of other RGC guidance cues netrin, DCC, and L1. Unexpectedly, errors occur in dorsal but not ventral retina despite early uniform or later high ventral expression of EphB2 and EphB3. Furthermore, embryos lacking EphB3 and the kinase domain of EphB2 do not show increased errors, consistent with a guidance role for the EphB2 extracellular domain. Thus, while Eph kinase function is involved in RGC axon mapping in the brain, RGC axon pathfinding within the retina is partially mediated by EphB receptors acting in a kinase-independent manner (Birgbauer, 2000).
A role of Eph receptor molecules independent of their kinase domain for pathfinding within the retina is in contrast to the proposed kinase-dependent function of Eph receptors in mediating RGC axon mapping onto CNS targets. However, since retinotopic mapping at the target has thus far only been demonstrated with EphA receptors, and the results here involve EphB receptors, it is formally possible that a difference exists such that EphA receptor function depends on kinase activity while EphB function is kinase independent. However, another possibility is that Eph receptors mediate RGC axon pathfinding at multiple sites along the visual pathway by different mechanisms. Thus, Eph function in the retina itself may be kinase independent, but Eph kinase activity may be essential later during RGC axon topographic mapping within CNS targets. It may be possible to differentiate between these two models by examining the role of EphA receptors in intra-retinal axon pathfinding and the involvement of the kinase domain of EphB receptors in RGC axon mapping within the superior colliculus (Birgbauer, 2000).
Axon pathfinding relies on cellular signaling mediated by growth cone receptor proteins responding to ligands, or guidance cues, in the environment. Eph proteins are a family of receptor tyrosine kinases that govern axon pathway development, including retinal axon projections to CNS targets. Recent examination of EphB mutant mice, however, has shown that axon pathfinding within the retina to the optic disc is dependent on EphB receptors, but independent of their kinase activity. This study shows a function for EphB1, B2 and B3 receptor extracellular domains (ECDs) in inhibiting mouse retinal axons when presented either as substratum-bound proteins or as soluble proteins directly applied to growth cones via micropipettes. In substratum choice assays, retinal axons have tended to avoid EphB-ECDs, while time-lapse microscopy shows that exposure to soluble EphB-ECD leads to growth cone collapse or other inhibitory responses. These results demonstrate that, in addition to the conventional role of Eph proteins signaling as receptors, EphB receptor ECDs can also function in the opposite role as guidance cues to alter axon behavior. Furthermore, the data support a model in which dorsal retinal ganglion cell axons heading to the optic disc encounter a gradient of inhibitory EphB proteins which helps maintain tight axon fasciculation and prevents aberrant axon growth into ventral retina. In conclusion, development of neuronal connectivity may involve the combined activity of Eph proteins serving as guidance receptors and as axon guidance cues (Birgbauer, 2001).
In Xenopus tadpoles, all retinal ganglion cells (RGCs) send axons contralaterally across the optic chiasm. At metamorphosis, a subpopulation of EphB-expressing RGCs in the ventrotemporal retina begin to project ipsilaterally. However, when these metamorphic RGCs are grafted into embryos, they project contralaterally, suggesting that the embryonic chiasm lacks signals that guide axons ipsilaterally. Ephrin-B is expressed discretely at the chiasm of metamorphic but not premetamorphic Xenopus. When expressed prematurely in the embryonic chiasm, ephrin-B causes precocious ipsilateral projections from the EphB-expressing RGCs. Ephrin-B is also found in the chiasm of mammals, which have ipsilateral projections, but not in the chiasm of fish and birds, which do not. These results suggest that ephrin-B/EphB interactions play a key role in the sorting of axons at the vertebrate chiasm (Nakagawa, 2000).
Ephrin-A2 and -A5 are thought to be anteroposterior mapping labels for the retinal axonal projection to the tectum or to its mammalian equivalent, the superior colliculus (SC). In this paper, gene disruptions for both these ephrins are characterized. Focal retinal labeling reveals moderate map abnormalities when either gene is disrupted. Double heterozygotes also have a phenotype, showing an influence of absolute levels. In vitro assays indicate these ephrins are required for repellent activity in the target and also normal responsiveness in the retina. In double homozygotes, anteroposterior order is almost though not completely lost. Temporal or nasal retinal labelings reveal quantitatively similar but opposite shifts, with multiple terminations scattered widely over the target. These results indicate an axon competition mechanism for mapping, with a critical role for ephrins as anteroposterior topographic labels. Dorsoventral topography is also impaired, showing these ephrins are required in mapping both axes (Feldheim, 2000).
Currently, a leading model is that both attractant and repellent gradients may be prespecified within the target. Each axon then identifies its correct termination zone as the point where repellent and attractant forces cancel out. In this model, wild-type nasal axons map to posterior SC because they are attracted there. A prediction of this model is that in the ephrin-A2-/-; ephrin-A5-/- double mutant, axons would tend to map more posteriorly than normal, or in the case of extreme nasal axons might show no shift. This appears to be contrary to the results obtained in this study, especially considering that the opposite shifts in the nasal and temporal axons are of similar magnitudes. For similar reasons these results do not seem consistent with a variant of the counterbalanced gradient model, where a posterior to anterior ephrin gradient would be balanced by a prespecified repellent in an anterior to posterior gradient. An alternative possibility that could perhaps be reconciled with these results would be for ephrins in the target to act as either repellents or attractants at different concentrations, although there is currently no direct evidence to support such an effect from axon guidance assays (Feldheim, 2000 and references therein).
As an alternative to models with two counterbalanced gradients prespecified in the target, these results could be explained by a model involving a repellent gradient of ephrins, in combination with axon-axon competition. This competition could be for limiting positive factors in the target or could also involve direct axon-axon interactions. A model of this type has been suggested in an analysis of the LGN in ephrin-A5-/- mice, and the model now receives strong support from the more comprehensive analysis in the SC presented here. The repulsion/competition model would account for normal mapping as follows. Temporal axons would be unable to terminate in the posterior SC, because they are repelled by the ephrins, so they would be forced to arborize in the anterior SC. Nasal axons are less sensitive to ephrin repulsion and so would be able to terminate in the posterior SC. In the anterior SC, nasal axons face greater competition for limiting amounts of permissive factor(s), so they prefer to avoid this competition and arborize only in the posterior SC (Feldheim, 2000).
Incorporating competition in the model can explain several aspects of the data. (1) Nasal axons shift anteriorly in the ephrin-A5-/- single mutant (although this could be explained by the axon sensitivity model outlined above), and shift even further to the anterior in the ephrin-A2-/-; ephrin-A5-/- double mutant. According to the competition model, posterior repellents are removed, so axons from temporal or central retina are now able to compete more effectively in posterior SC; nasal axons therefore face increased competition in the posterior SC, so they lose their strong preference for this region and spread out into more anterior regions. The model thus seems to fit well with the opposite and quantitatively similar shifts of axons from temporal and nasal extremes of the retina. Labelings at intermediate retinal positions are more difficult to characterize because of the lack of fixed landmarks but, consistent with the competition model, such labelings revealed multiple arborizations, seemingly shifted in both anterior and posterior directions. (2) Axons are not respecified to a specific ectopic position. Instead, arborizations are scattered over an abnormally broad zone in the mutants, including both normal and abnormal regions. This is the result predicted by the competition model: as the topographically specific repellents are removed, axons would spread into abnormal regions, but there is no reason for them to disfavor the correct region. (3) Even in the ephrin-A2-/-; ephrin-A5-/- mutant, retinal axons fill the entire SC and axons from both nasal and temporal extremes of the retina form connections within the target. Models that involve a strict matching of values in the projecting and target field would predict that the mutant should have unmatched areas, whereas the competition model predicts that the entire projecting and target fields should still match up (Feldheim, 2000).
In the mammalian visual system, retinal axons undergo temporal and spatial rearrangements as they project bilaterally to targets on the brain. Retinal axons cross the neuraxis to form the optic chiasm on the hypothalamus in a position defined by overlapping domains of regulatory gene expression. However, the downstream molecules that direct these processes remain largely unknown. A novel in vitro paradigm was used to study possible roles of the Eph family of receptor tyrosine kinases in chiasm formation. In vivo, Eph receptors and their ligands distribute in complex patterns in the retina and hypothalamus. In vitro, retinal axons are inhibited by reaggregates of isolated hypothalamic, but not dorsal diencephalic or cerebellar cells. Furthermore, temporal retinal neurites are more inhibited than nasal neurites by hypothalamic cells. Addition of soluble EphA5-Fc to block Eph 'A' subclass interactions decreases both the inhibition and the differential response of retinal neurites by hypothalamic reaggregates. These data show that isolated hypothalamic cells elicit specific, position-dependent inhibitory responses from retinal neurites in culture. Moreover, these responses are mediated, in part, by Eph interactions. Together with the in vivo distributions, these data suggest possible roles for Eph family members in directing retinal axon growth and/or reorganization during optic chiasm formation (Marcus, 2000).
Repulsion plays a fundamental role in the establishment of a topographic map of the chick retinotectal projections. This has been highlighted by studies demonstrating the role of opposing gradients of the EphA3 receptor tyrosine kinase on retinal axons and two of its ligands, ephrin-A2 and ephrin-A5, in the tectum. The distribution of these two ephrins in other retinorecipient structures in the chick diencephalon and mesencephalon has been examined during the period when visual connections are being established. Both ephrin-A2 and ephrin-A5 and their receptors EphA4 and EphA7 are expressed in gradients whose orientation is consistent with the topography of the nasotemporal axis of the respective retinofugal projections. In addition, their distribution suggests that receptor-ligand interactions may be involved in the organization of connections between the different primary visual centers and, thus, in the topographic organization of secondary visual projections. Interestingly, where projections lack a clear topographic representation, a uniform expression of the Eph-ephrin molecules is observed. A similar patterning mechanism may be implicated in the transfer of visual information to the telencephalon. These results suggest a conserved function for EphA receptors and their ligands in the elaboration of topographic maps at multiple levels of the visual pathway (Marin, 2001).
The idea has been put forward that molecules and mechanisms acting during development are re-used during regeneration in the adult, for example in response to traumatic injury in order to re-establish the functional integrity of neuronal circuits. Members of the Eph family of receptor tyrosine kinases and their 'ligands', the ephrins, play a prominent role during development of the retinocollicular projection in rodents, where EphA receptors and ephrin-As are expressed in gradients in both the retina and the superior colliculi (SC). Whether EphA family members are also expressed or re-expressed in the adult after optic nerve lesion was examined, since the presence of axon guidance information is an important prerequisite for a topographically appropriate re-connection by retinal ganglion cell (RGC) axons. This analysis was encouraged by results showing that RGC axons do not exert guidance preferences in response to membranes from adult unlesioned SC, but in response to membranes from the adult deafferented SC. A graded expression pattern of ephrin-As was found in the SC both before and after deafferentation, which is remarkably similar to those found during development. EphA receptor levels are reduced in the SC after deafferentation and the expression patterns of the EphB family are not changed. In particular, the presence of a graded ephrin-A expression in the deafferented SC suggests that -- if robust regeneration of RGC axons can be achieved -- topographic guidance information as a likely requirement for a functionally successful re-establishment of the retinocollicular projection is available (Knoll, 2001b).
The retinotectal projection is the predominant model for studying molecular mechanisms controlling development of topographic axonal connections. Analyses of topographic mapping of retinal ganglion cell (RGC) axons in chick optic tectum indicate that a primary role for guidance molecules is to regulate topographic branching along RGC axons, a process that imposes unique requirements on the molecular control of map development. Topographically appropriate connections are established exclusively by branches that form along the axon shaft. Initially, RGC axons overshoot their appropriate termination zone (TZ) along the anterior-posterior (A-P) tectal axis; temporal axons overshoot the greatest distance and nasal axons the least: this correlates with the nonlinear increasing A-P gradient of ephrin-A repellents. In contrast, branches form along the shaft of RGC axons with substantial A-P topographic specificity. Topography is enhanced through the preferential arborization of appropriately positioned branches and elimination of ectopic branches. Using a membrane stripe assay and time-lapse microscopy, it has been shown that branches form de novo along retinal axons. Temporal axons preferentially branch on their topographically appropriate anterior tectal membranes. After the addition of soluble EphA3-Fc, which blocks ephrin-A function, temporal axons branch equally on anterior and posterior tectal membranes, indicating that the level of ephrin-As in posterior tectum is sufficient to inhibit temporal axon branching and generate branching specificity in vitro. These findings indicate that topographic branch formation and arborization along RGC axons are critical events in retinotectal mapping. Ephrin-As inhibit branching along RGC axons posterior to their correct TZ, but alone cannot account for topographic branching and must cooperate with other molecular activities to generate appropriate mapping along the A-P tectal axis (Yates, 2001).
In animals with binocular vision, retinal ganglion cell (RGC) axons either cross or avoid the midline at the optic chiasm. Ephrin-Bs in the chiasm region direct the divergence of retinal axons through the selective repulsion of a subset of RGCs that express EphB1. Ephrin-B2 is expressed in radial glia at the mouse chiasm midline as the ipsilateral projection is generated and is selectively inhibitory to axons from ventrotemporal (VT) retina, where ipsilaterally projecting RGCs reside. Moreover, blocking ephrin-B2 function in vitro rescues the inhibitory effect of chiasm cells and eliminates the ipsilateral projection in the semiintact mouse visual system. A receptor for ephrin-B2, EphB1, is found exclusively in regions of retina that give rise to the ipsilateral projection. EphB1 null mice exhibit a dramatically reduced ipsilateral projection, suggesting that this receptor contributes to the formation of the ipsilateral retinal projection, most likely through its repulsive interaction with ephrin-B2. This study provides the first direct evidence that ephrin-B2 is more inhibitory to ipsilateral than contralateral retinal axons and demonstrates that ephrin-B2 is necessary for the ipsilateral projection to form. Using in situ hybridization and analyses of mutant mice, EphB1, a receptor for ephrin-B2, is shown to be expressed specifically in regions of the retina that give rise to the ipsilateral projection, and in mice lacking EphB1. Together these data identify ephrin-B2 and EphB1 as key players in retinal axon divergence and suggest that the function of B-class Ephs and ephrins in patterning binocular vision is conserved between species (Williams, 2003).
During development of the retinocollicular projection in mouse, retinal axons initially overshoot their future termination zones (TZs) in the superior colliculus (SC). The formation of TZs is initiated by interstitial branching at topographically appropriate positions. Ephrin-As are expressed in a decreasing posterior-to-anterior gradient in the SC, and they suppress branching posterior to future TZs. This study investigates the role of an EphA7 gradient in the SC, which has the reverse orientation compared to the ephrin-A gradient. In EphA7 mutant mice the retinocollicular map is disrupted, with nasal and temporal axons forming additional or extended TZs, respectively. In vitro, retinal axons are repelled from growing on EphA7-containing stripes. These data support the idea that EphA7 is involved in suppressing branching anterior to future TZs. These findings suggest that opposing ephrin-A and EphA gradients are required for the proper development of the retinocollicular projection (Rashid, 2005).
The cell surface proteoglycan syndecan-2 can induce dendritic spine formation in hippocampal neurons. The EphB2 receptor tyrosine kinase phosphorylates syndecan-2 and this phosphorylation event is crucial for syndecan-2 clustering and spine formation. Syndecan-2 is tyrosine phosphorylated and forms a complex with EphB2 in mouse brain. Dominant-negative inhibition of endogenous EphB receptor activities blocks clustering of endogenous syndecan-2 and normal spine formation in cultured hippocampal neurons. This is the first evidence that Eph receptors play a physiological role in dendritic spine morphogenesis. These observations suggest that spine morphogenesis is triggered by the activation of Eph receptors: this causes tyrosine phosphorylation of target molecules, such as syndecan-2, in presumptive spines (Ethell, 2001).
Dendritic spines are the principal postsynaptic targets for excitatory synapses. In recent years, these small protrusions on the surface of dendrites have attracted significant interest because changes in their morphology are implicated in synaptic plasticity and long-term memory. Dendritic spines undergo morphological changes in a developmentally regulated and activity-dependent manner. Abnormal spine morphologies have been reported in several neurodevelopmental disorders, including fragile X syndrome. The molecular mechanisms that govern spine morphogenesis are not completely understood, but several different physiological and molecular factors have been shown to affect spine morphology. These factors include synaptic activity and plasticity, actin filament reorganization, calcium dynamics, and protein phosphorylation (Ethell, 2001 and references therein).
Dendritic spine formation occurs during the late stages of development after neuronal connectivity has been established. Before the appearance of mature spines, dendrites exhibit long, thin filopodia-like protrusions without a bulbous head. As the brain matures, these dendritic filopodia disappear, and spines, which typically have mushroom-like and stubby shapes, begin to appear. Primary cultures of rat hippocampal neurons provide an excellent system in which the process of spine formation can be studied in vitro. At 1 week in vitro, these neurons possess predominantly filopodia-like protrusions. Over the next few weeks, these dendritic filopodia gradually decrease in number and are progressively replaced by protrusions that have mushroom-like and stubby shapes. After 3-4 weeks, the majority of the protrusions exhibit mature spine morphologies. The cell surface proteoglycan syndecan-2 has been shown to play a role in spine formation (Ethell, 2001 and references therein).
Syndecan-2 is a member of the syndecan family of transmembrane heparan sulfate proteoglycans. There is increasing evidence that syndecans are involved in transmembrane signaling by interacting with cytoskeletal and signaling molecules. The cytoplasmic domain of syndecans can be subdivided into a highly conserved juxtamembrane segment (C1 region), another conserved segment at the C terminus (C2 region), and a variable segment (V region) located between the C1 and C2 regions. The EFYA (Glu-Phe-Tyr-Ala) sequence at the C terminus serves as the binding site for at least four cytoplasmic proteins, namely syntenin, CASK, synectin, and synbindin. Moreover, the cytoplasmic domain contains 4 tyrosine residues that are conserved among all syndecans. Some of these tyrosine residues are phosphorylated in vitro, and tyrosine phosphorylation has been speculated to play important roles in syndecan-mediated signal transduction (Ethell, 2001 and references therein).
Potential functional roles of syndecan-2 in synapses were first suggested by its interaction with the synaptic PDZ domain protein CASK. Syndecan-2 is clustered at dendritic spines of mature hippocampal neurons in culture and its accumulation occurs concomitant with the morphological maturation of spines. More importantly, transfection of syndecan-2 induces the formation of morphologically mature dendritic spines in immature (8 days in vitro [DIV]) hippocampal neurons. Deletion studies have demonstrated that the C1 and V regions of the syndecan-2 cytoplasmic domain (which contains two potential tyrosine phosphorylation sites) are required for syndecan-2 clustering on dendrites and the induction of mature spines. Based on these data, it is hypothesized that tyrosine phosphorylation of syndecan-2 is the crucial upstream event that leads to dendritic spine formation. This premise then suggests that tyrosine kinase(s) present in dendritic spines play a role in spine formation by phosphorylating syndecan-2 (Ethell, 2001 and references therein).
The Eph family is a large family of receptor tyrosine kinases. Upon stimulation by ephrin ligands, Eph receptors activate signaling cascades in various biological systems. While Eph receptors have been studied primarily in the context of axon guidance during development, there have been suggestions that they may play some roles in synapses in the adult brain. It is speculated that Eph receptors are the kinases involved in the syndecan-2-induced spine formation for several reasons.: (1) syndecan-3, another member of the syndecan family, is tyrosine phosphorylated by recombinant EphB1 in vitro; (2) some Eph receptors interact with syntenin, a syndecan-2 binding PDZ domain protein, and (3) most importantly, some Eph receptors, including EphB2, are present in dendritic spines. These observations have led to this investigation of the possibility that Eph receptors are involved in syndecan-2 phosphorylation during dendritic spine formation (Ethell, 2001 and references therein).
In this paper, it is demonstrated that EphB2 is a crucial tyrosine kinase that phosphorylates syndecan-2 during dendritic spine formation. Furthermore, inhibition of endogenous EphB receptor activities by dominant-negative EphB2 blocks endogenous syndecan-2 clustering and normal spine formation. These results demonstrate a physiological role for EphB2/syndecan-2 signaling in dendritic spine morphogenesis. These findings provide a basis for the role of cell surface ligand-receptor interactions in spine morphogenesis and suggest that the signaling cascade leading to the formation of mature spines is triggered by the activation of Eph receptors by their extracellular ligands (Ethell, 2001).
The morphogenesis of dendritic spines, the major sites of excitatory synaptic transmission in the brain, is important in synaptic development and plasticity. An ephrinB-EphB receptor trans-synaptic signaling pathway has been identified that regulates the morphogenesis and maturation of dendritic spines in hippocampal neurons. Activation of the EphB receptor induces translocation of the Rho-GEF kalirin (Drosophila ortholog: Trio) to synapses and activation of Rac1 and its effector PAK. Overexpression of dominant-negative EphB receptor, catalytically inactive kalirin or dominant-negative Rac1, or inhibition of PAK each eliminates ephrin-induced spine development. This novel signal transduction pathway may be critical for the regulation of the actin cytoskeleton controlling spine morphogenesis during development and plasticity (Penzes, 2003).
The role of the Rac1 effector p21-activated kinase PAK was examined. Several PAK proteins are expressed in the brain, and previous studies have shown that some of the effects of Rac1 on the cytoskeleton are mediated by PAK. In addition, genetic analysis in Drosophila has shown that PAK1 is genetically associated with Trio, the fly ortholog of kalirin, in the pathway through which Trio affects axon growth and guidance. Binding of activated Rac1 to PAK induces PAK autophosphorylation, which strongly correlates with its activation. To test whether ephrinB treatment induces activation of PAKs, an antibody detecting autophosphorylated PAK (P-PAK) was used. In addition, this experiment can be regarded as a way to visualize endogenous Rac1 activation. Treatment of hippocampal neurons with clustered ephrinB1 induce a dramatic increase in the number and size of clusters stained with the P-PAK antibody. This effect was confirmed by Western analysis with the P-PAK antibody of extracts of 4-week-old high-density cortical neurons treated with ephrinB1. Moreover, in hippocampal neurons, ephrinB1 treatment induces activation of PAK at synapses, as shown by P-PAK immunostaining coincident with synaptophysin (Penzes, 2003).
To test whether kalirin-7 was required for ephrinB1-induced PAK phosphorylation, the effect was examined of overexpressing the GEF inactive kal7-mut in hippocampal neurons on the ability of clustered ephrinB1 to induce phosphorylation of PAK. Therefore, DIV7 hippocampal neurons were transfected with myc-kal7-mut, and 2 days later the neurons were treated with clustered ephrinB1 for 2 hr, followed by fixation and immunostaining for myc and P-PAK. While ephrinB1 treatment induces an increased phosphorylation of PAK in nontransfected neurons, in neurons expressing kal7-mut, the level of P-PAK is visibly reduced compared to adjacent nontransfected neurons. Quantification of the ratios of P-PAK fluorescence intensities to total cell areas of nontransfected control neurons relative to the same ratios for neurons expressing myc-kal7-mut confirmed this observation (Penzes, 2003).
PAKs phosphorylate proteins involved in regulating the actin cytoskeleton and gene expression. To test whether PAK is an essential downstream component of ephrinB signaling in spine morphogenesis, GFP-transfected hippocampal neurons were treated with a fusion protein of the PAK1 inhibitory domain (PID) fused with the cell-penetrating peptide (TAT-PID) along with ephrinB1. These neurons exhibit a reduction in the number and size of spines, compared to the ephrinB1-treated neurons, while also showing a reduced phosphorylation level of PAK, confirming its inhibition by PID. Together, these data demonstrate that Rac1 and PAK are key downstream components of ephrinB regulation of spine morphogenesis (Penzes, 2003).
During development, it is necessary to coordinate accurately the formation and location of presynaptic active zones with those of the postsynaptic structures. This could be achieved by signaling from presynaptic ephrinB, clustered at active zones on axons, to activate postsynaptic EphB2, resulting in synaptogenesis on the apposing dendrites. Even in mature neurons, dendritic spines are very dynamic structures, and recent studies have demonstrated that LTP induces morphological changes in spines, which may contribute to plasticity in adult neurons. The rapid and dramatic effect of ephrinB on spine maturation suggests that ephrinB-EphB2 signaling may be a key component in the regulation of spine morphogenesis during plasticity. Other extracellular signals have been shown to regulate spine morphogenesis, such as K+ depolarization, glutamate action on NMDA receptors, and BDNF. It is possible that kalirin mediates the intracellular effects of these signals as well (Penzes, 2003).
Pattern formation in the hindbrain involves a segmentation process leading to the formation of metameric units, manifested as successive swellings known as rhombomeres (r). In search for genes involved in cell-cell interactions during hindbrain segmentation, a screen was carried out for protein kinase genes with restricted expression patterns in this region of the CNS. Three novel mouse genes, Sek-2, Sek-3 and Sek-4 (members of the Eph subfamily of putative transmembrane receptor protein tyrosine kinases [RTKs]) were cloned, their chromosomal locations were identified, and their expression between 7.5 and 10.5 days of development was analyzed. Before morphological segmentation, Sek-2 is transcribed in a transverse stripe corresponding to prospective r4 and the adjacent mesoderm, suggesting possible roles both in hindbrain segmentation and signaling between neuroepithelium and mesoderm. Sek-3 and Sek-4 have common domains of expression, including r3, r5 and part of the midbrain, as well as specific domains in the diencephalon, telencephalon, spinal cord and in mesodermal and neural crest derivatives. Together with the finding that Sek (Sek-1) is expressed in r3 and r5, these data indicate that members of the Eph family of RTKs may co-operate in the segmental patterning of the hindbrain (Becker, 1994).
Described here is the isolation and characterization of two zebrafish Eph receptor ligand cDNAs that have been called zfEphL3 and zfEphL4. These genes are expressed in the presumptive midbrain of developing embryos from 6 somites. By 24 hours, L3 is expressed throughout the midbrain including the region of the presumptive tectum, whereas L4 is strongly expressed in the midbrain caudal to the presumptive tectum. At later stages of development L3 is expressed in a graded fashion throughout the tectum and L4 is maintained at its posterior margin. Growth cone collapse and pathway selection assays demonstrate that both these proteins have a collapse activity for retinal ganglion cells. When faced with a choice of substrate on which to grow, temporal axons from chick retinal ganglion cells selectively avoid membranes from Cos cells transfected with L3, whereas nasal axons do not. Both temporal and nasal axons avoid membranes from Cos cells transfected with L4. The expression patterns, together with the functional data, suggest that although both ligands may be able to guide retinal ganglion cells axons in vitro, they have different roles in the guidance of retinotectal projections in vivo. The expression of L3 is consistent with a role in the guidance of retinal ganglion cells to their targets on the tectum, whereas that of L4 suggests a role in delineating the posterior boundary of the optic tectum (Brennan, 1997).
Segmentation of the vertebrate hindbrain leads to the formation of a series of rhombomeres (r) with distinct identities. Recent studies have uncovered regulatory links between transcription factors governing this process, but little is known of how these relate to molecules mediating cell-cell signaling. The Eph receptor tyrosine kinase gene EphA4 (Sek-1) is expressed in r3 and r5, and function-blocking experiments suggest that it is involved in restricting intermingling of cells between odd- and even-numbered rhombomeres. The cis-acting regulatory sequences of the EphA4 gene have been analyzed in transgenic mice and a 470 bp enhancer element that drives specific expression in r3 and r5 has been identified. Within this element, eight binding sites have been identified for the Krox-20 transcription factor that is also expressed in r3 and r5. Mutation of these binding sites abolishes r3/r5 enhancer activity and ectopic expression of Krox-20 leads to ectopic activation of the enhancer. These data indicate that Krox-20 is a direct transcriptional activator of EphA4. Together with evidence that Krox-20 regulates Hox gene expression, these findings reveal a mechanism by which the identity and movement of cells are coupled such that sharply restricted segmental domains are generated (Theil, 1998).
The restriction of intermingling between specific cell populations is crucial for the maintenance of organized patterns during development. A striking example is the restriction of cell mixing between segments in the insect epidermis and the vertebrate hindbrain that may enable each segment to maintain a distinct identity. In the hindbrain, this is a result of different adhesive properties of odd- and even-numbered segments (rhombomeres), but an adhesion molecule with alternating segmental expression has not been found. However, blocking experiments suggest that Eph-receptor tyrosine kinases may be required for the segmental restriction of cells. Eph receptors and their membrane-bound ligands, ephrins, are expressed in complementary rhombomeres and, by analogy with their roles in axon pathfinding, could mediate cell repulsion at boundaries. Remarkably, transmembrane ephrins can themselves transduce signals, raising the possibility that bi-directional signaling occurs between adjacent ephrin- and Eph-receptor-expressing cells. Mosaic activation of Eph receptors leads to sorting of cells to boundaries in odd-numbered rhombomeres, whereas mosaic activation of ephrins results in sorting to boundaries in even-numbered rhombomeres. These data implicate Eph receptors and ephrins in the segmental restriction of cell intermingling (Xu, 1999).
Rhombomeres are segmental units of the developing vertebrate hindbrain that underlie the reiterated organization of cranial neural crest migration and neuronal differentiation. valentino (val), a zebrafish homolog of the mouse bzip transcription factor-encoding gene, kreisler (potential Drosophila homolog: CG10034), is required for segment boundary formation caudal to rhombomere 4 (r4). val is normally expressed in r5/6 and is required for cells to contribute to this region. In val minus mutants, rX, a region one rhombomere in length and of mixed identity, lies between r4 and r7. While a number of genes involved in establishing rhombomeric identity are known, it is still largely unclear how segmental integrity is established and boundaries are formed. Members of the Eph family of receptor tyrosine kinases and their ligands, the ephrins, are candidates for functioning in rhombomere boundary formation. Indeed, expression of the receptor ephB4a coincides with val in r5/6, while ephrin-B2a, which encodes a ligand for EphB4a, is expressed in r4 and r7, complementary to the domain of val expression (Cooke, 2001).
In val minus embryos, ephB4a expression is downregulated and ephrin-B2a expression is upregulated between r4 and r7, indicating that Val is normally required to establish the mutually exclusive expression domains of these two genes. Juxtaposition of ephB4a-expressing cells and ephrin-B2a-expressing cells in the hindbrain leads to boundary formation. Loss of the normal spatial regulation of eph/ephrin expression in val mutants correlates not only with absence of boundaries but also with the inability of mutant cells to contribute to wild-type r5/6. Using a genetic mosaic approach, it has been shown that spatially inappropriate Eph signaling underlies the repulsion of val minus cells from r5/6. It is proposed that Val controls eph expression and that interactions between EphB4a and Ephrin-B2a mediate cell sorting and boundary formation in the segmenting caudal hindbrain (Cooke, 2001).
Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.
The Interactive Fly resides on the
Society for Developmental Biology's Web server.