Contralateral migration of oculomotor neurons is regulated by Slit/Robo signaling
© The Author(s). 2016
Received: 12 June 2016
Accepted: 11 October 2016
Published: 22 October 2016
Oculomotor neurons develop initially like typical motor neurons, projecting axons out of the ventral midbrain to their ipsilateral targets, the extraocular muscles. However, in all vertebrates, after the oculomotor nerve (nIII) has reached the extraocular muscle primordia, the cell bodies that innervate the superior rectus migrate to join the contralateral nucleus. This motor neuron migration represents a unique strategy to form a contralateral motor projection. Whether migration is guided by diffusible cues remains unknown.
We examined the role of Slit chemorepellent signals in contralateral oculomotor migration by analyzing mutant mouse embryos.
We found that the ventral midbrain expresses high levels of both Slit1 and 2, and that oculomotor neurons express the repellent Slit receptors Robo1 and Robo2. Therefore, Slit signals are in a position to influence the migration of oculomotor neurons. In Slit 1/2 or Robo1/2 double mutant embryos, motor neuron cell bodies migrated into the ventral midbrain on E10.5, three days prior to normal migration. These early migrating neurons had leading projections into and across the floor plate. In contrast to the double mutants, embryos which were mutant for single Slit or Robo genes did not have premature migration or outgrowth on E10.5, demonstrating a cooperative requirement of Slit1 and 2, as well as Robo1 and 2. To test how Slit/Robo midline repulsion is modulated, we found that the normal migration did not require the receptors Robo3 and CXCR4, or the chemoattractant, Netrin 1. The signal to initiate contralateral migration is likely autonomous to the midbrain because oculomotor neurons migrate in embryos that lack either nerve outgrowth or extraocular muscles, or in cultured midbrains that lacked peripheral tissue.
Overall, our results demonstrate that a migratory subset of motor neurons respond to floor plate-derived Slit repulsion to properly control the timing of contralateral migration.
KeywordsOculomotor Motor neuron Migration Floor plate Slit/Robo
The oculomotor neurons are the most anterior motor neurons in the CNS, forming the oculomotor nerve (nIII). Their axons emerge from the ventral midbrain and innervate four of the six extraocular muscles. Development of the oculomotor system must occur with spatial and temporal accuracy to properly position the motor neuron cell bodies and to guide their axons to corresponding extraocular muscles. Errors in development can lead to abnormal eye movements or alignment, termed strabismus, and may result in partial blindness reviewed in . The mechanisms that guide the development of the oculomotor system remain poorly understood.
Early in embryonic development, clusters of oculomotor neurons project axons ipsilaterally toward muscle targets, similar to most motor neurons. However, during extraocular muscle innervation, a subset of neurons in the oculomotor nucleus repolarize to send a second process into and across the midline. This subset of motor neuron cell bodies then migrate across the ventral midbrain with axons trailing to join the contralateral oculomotor nucleus [2–7]. This process generates the oculomotor commissure and connects motor neurons located in the caudal half of the oculomotor nucleus to the contralateral superior (dorsal) rectus muscle. Contralateral innervation of the superior rectus muscle is highly conserved among vertebrates [8, and references within].
Oculomotor neurons navigate across the embryonic midline independent of an identifiable glial scaffold, and it was suggested that a “diffusible substance” guides the migrating neurons across the midline . The embryonic ventral midline, consisting of specialized floor plate tissue, is a source of diffusible guidance factors . However, it is unknown whether floor plate guidance cues guide oculomotor neurons. We have focused on two diffusible guidance factors that regulate midline crossing of commissural axons, the Slit proteins and Netrin1. In the developing spinal cord and hindbrain, Slits and Netrin1 mediate migration across the floor plate by their opposing chemotactic actions. The Slit proteins repel both navigating axons and migrating neuron cell bodies reviewed in [10, 11]. In vertebrates, there are three Slit proteins, of which Slit1 and 2 act at the midline to repel axons that express the Slit receptor Robo1 or 2 [12–14]. The third Slit receptor, Robo3 may either counteract the repellent activity of Robo1 and 2  or facilitate Netrin1 attraction . In contrast with Slit signaling, Netrin1 attracts both axons and neurons toward the midline thorough the receptor Deleted in Colorectal Cancer (DCC) [17–21]. Importantly, prior studies in hindbrain showed that these midline guidance signals are important for positioning other cranial neurons, including cranial axon repulsion by Netrin1 , and facial branchiomotor migration defects in Slit and Robo mutants and Netrin mutants [23, 24]. Both Slits and Netrin1 are expressed at the ventral midline in the midbrain during early developmental stages [25, 26]. The expression of both Slits and Netrin1 in the developing midbrain, coupled with their role in guiding midline crossing of axons, suggests a role for these cues in guiding the midline migration of oculomotor neurons.
Here we describe how Slit, but not Netrin, acts to gate migration of oculomotor cell bodies into the floor plate. We show that the initial clusters of motor neurons are in fact not static, but have considerable migratory potential, which is demonstrated by abnormal early migration across the floor plate when Slit/Robo signaling is disrupted.
Animal experiments were approved by the UNR IACUC, following NIH guidelines, with the approved protocol #2015-00435. DCC embryos are previously described . Robo, Slit and Netrin1 mutant mice were a kind gift from Marc Tessier-Lavigne (Rockefeller), and Frederic Charron (ICMR, Montreal CA). Mating to obtain various Slit mutant combinations was previously described ). Robo, Slit and Netrin1 PCR genotyping were performed as previously described [12, 27–29]. CXCR4 mutant embryos were a gift from John Rubenstein (UCSF). Pitx2 mutant embryos were a kind gift from Philip J Gage (University of Michigan, Ann Arbor). Images of Robo3 mutant embryos were provided by Alain Chedotal (INSERM, Paris). Wild type CD-1 mice were purchased from Charles River Laboratories (Wilmington, MA USA). Embryos were collected in the afternoon of day 10.5, 13.5, or 16.5 with embryonic day 0.5 designated as the day of the vaginal plug. Embryos were fixed with 4 % paraformaldehyde (PFA) overnight or for several days. E16.5 embryos were fixed via cardiac perfusion, and fixed overnight in 4 % PFA.
In situ hybridization
Whole mount in situ hybridization was previously described . Probes for Slit1,Slit2, Slit3, Robo1, Robo2, Robo3 were a kind gift from Marc Tessier-Lavigne, (Rockefeller).
CD1 E10.5 and E13.5 embryos were dissected in cold 0.4 M phosphate buffer, and fixed with 0.4 % PFA for 1 day. Embryos were then embedded for cryostat sectioning as described . 20 um cryostat sections were rinsed in warm 0.4 % phosphate buffer, and washed with PBS with 0.1 % TritonX-100, and 10 % normal goat serum (PBST). Primary antibodies were diluted in PBST and were applied overnight at room temperature. Primary antibodies included Robo3 (anti-rabbit, Abcam) 1:200, β-galactosidase (Jackson) 1:10,000, Islet1/2 ( DHSB, 39.4D5) 1:200, Robo1 and Robo2 antibodies (kind gift from Elke Stein, Yale; validated in ) 1:10,000. Sections were washed several times in PBST and Secondary antibody was applied. Secondary antibodies (Alexa 488, Alexa 555) were diluted in PBST and used at 1:200 for one hour at room temperature. For whole mount labeling of Islet1/2, tissue was placed in primary antibody for 4 days diluted in PBST. The biotin-avidin system (Invitrogen) was used for Islet1/2 amplification, biotin (donkey anti-mouse, 1:100 in PBST) was applied overnight at 4°, washed overnight in PBST, and followed by avidin555 or 488 (1:200in PBST) overnight at 4°.
To back-label the oculomotor nucleus and midline crossing fibers, the lipophilic dye, DiI, red, or DiO, green, was crushed onto the oculomotor nerve. First, the skin and mesenchymal tissue was carefully removed from the cephalic flexure ventral to the midbrain to reveal the nIII nerve, then forceps were used to pinch a small crystal of dye onto the nerve. Embryos were placed in 4 % PFA with .1 % EDTA at 37 °C overnight (E10.5) or up to 3 days (E13.5, E16.5) to allow dye to travel. To visualize the labeled oculomotor nucleus, a 200 um coronal section was cut with a vibratome, and imaged with a confocal microscopy. To visualize the anterior-posterior axis of the OM nucleus, the embryo was cut along the dorsal edge to reveal the midline (open book preparation).
Quantification of motor neurons in the floorplate
E10.5 midbrains were dissected from various combinations of Slit and Robo mutant mice and antibody labeled for Islet1/2 in open book or sectioned preparations. Islet-positive cells that were located in the space between the two defined oculomotor nuclei were quantified. The average number of cells was graphed with standard deviation indicated by error bars. Significance was determined by Tukey HSD one-way-ANOVA.
Explant cultures of isolated midbrain tissue
E11.5 mouse midbrain tissues were dissected away from peripheral tissues as an open book preparation, and were cultured in a three-dimensional collagen gel matrix. The cultured tissues were fixed in 4 % PFA after 0, 24, 48, and 72 h of incubation. To label the migrated neurons, explant tissues were washed in PBS containing 10 % FBS and 1 % Triton for several hours (PBST). Primary antibody (1:200 mouse anti-Islet1/2, DSHB) in PBST was applied for 3-4 days. After washing the tissues for several hours, secondary antibody (1:200 Cy3 anti-rabbit (Invitrogen)) in PBST was applied for 2-3 days. The tissues were washed again and mounted on the slides for image acquisition under the confocal microscope (Olympus FV10-ASW).
A subset of oculomotor neurons migrate across the midline of the midbrain
To investigate contralateral migration of oculomotor neurons in wild type mice, we first determined the time course of normal migration. To distinguish cell bodies and projections that originate from the left or right oculomotor nuclei, we back-labeled the right and left nucleus with DiI and DiO crushed onto the right and left nIII, respectively. We refer to cells and axons labeled through nIII in this manner as oculomotor, although we note that the embryonic nIII also includes visceral motor fibers from the closely-associated Edinger-Westphal nucleus. To determine the location of oculomotor neuron cell bodies, we labeled the ventral midbrain with Islet1/2 antibody, a general motor neuron marker .
The Slit family of guidance cues and their receptors are in position to inhibit migration into the floor plate on E10.5
Slit1 and 2 expression was previously shown to be prominent in ventral midbrain in E12 rat embryos , and in E10.5 mouse embryos . To more specifically determine whether Slits are in a position to act as midline repellents prior to E13.5 in mouse, we examined the three Slits in the ventral midbrain on E10.5 by in situ hybridization of mRNA. We found Slit 1, 2 and 3 mRNA was expressed in the ventral floor plate medial to the oculomotor nuclei, as well as a strip of cells in the underlying ventricular zone (Fig. 3C-E) on E10.5. Complementary expression of Robo receptors by oculomotor neurons and Slit expression in the floor plate supports a role for Slit repulsion from the floor plate, and may act to inhibit oculomotor migration prior to E13.5. The ventricular zone expression of Slits suggests a potential role in hemming the neurons into the marginal zone. We also found Slit2 and 3 expression overlaps with the region of the Islet1/2-positive oculomotor neurons (Fig. 3D, E). This is consistent with Slit2 and 3 expression found in spinal motor neurons .
Oculomotor cell bodies migrate prematurely in Slit and Robo mutant mice
Loss of Slit 2 in motor neurons does not cause premature migration
The oculomotor commissure forms properly in Slit and Robo mutant mice
Oculomotor neuron migration across the midline on E10.5 in Slit and Robo mutant mice could be non-specifically affecting the entire nucleus, or could represent a premature but specific migration of those that normally migrate, that is, the superior rectus subset. Unfortunately, we were unable to identify a molecular marker for the superior rectus neurons in mouse embryos. However, we predicted that the oculomotor commissure would be larger if another population of cells migrated across the floor plate and joined the contralateral nucleus to augment the usual superior rectus commissural axon projection.
To determine if Slit and Robo influence the final development of the commissure on E16.5, DiI-labeled oculomotor nuclei were sectioned across a coronal plane through the commissure. In controls, the oculomotor commissure was fully developed with leading processes, spanning the floor plate (Fig. 8D). DiI-labeled processes intercalated throughout the contralateral nucleus. No cell bodies were seen within the floor plate, suggesting that the contralateral migration was complete by E16.5 (Fig. 8D). In E16.5 Slit1/2 and Robo1/2 mutants, the oculomotor commissure was positioned caudally, as in controls, with a similar size of the commissure (Fig. 8E, F). Together, this data suggests that premature migration on E10.5 does not appear to influence normal migration at later stages.
Regulators of Slit signaling are not required for contralateral migration
Migration of oculomotor cells on E10.5 in Slit and Robo mutant embryos suggests a mechanism in which wild type oculomotor neurons are initially trapped in position by Slit/Robo repulsion, but later, on E13.5, suppression of Slit/Robo signals allow for migration into and across the midline. We considered and tested three mechanisms that suppress Robo repulsive signaling in other systems.
A second potential antagonist of Slit signaling is the receptor CXCR4. In chick, the binding of SDF-1 to CXCR4 suppresses the repellent activity of Slit2, Semaphorin3A, and 3C on axons cultured in vitro . In zebrafish retinal ganglion cells, SDF-1 antagonizes Slit/ Robo2 signaling in vivo . CXCR4 is expressed in the mouse ventral midbrain where it is required for the exit of oculomotor and other cranial nerves . We confirmed CXCR4 expression in the oculomotor nucleus by in situ hybridization (Fig. 9B). Therefore, we were interested in whether floor plate CXCR4 may antagonize Slit signaling in migrating oculomotor neurons to allow for midline crossing. However, in E13.5 CXCR4 mutant embryos, Islet1/2-positive cells were found in the midline (Fig. 9D). This suggests that CXCR4 signaling is not required to antagonize the repellent activity of the floor plate.
Netrin1 attraction is not required for oculomotor migration into the floor plate
Contralateral OM migration is regulated by signals intrinsic to the ventral midbrain
The initiation of contralateral migration coincides with the time when the oculomotor axons reach the superior rectus extraocular muscle primordia in chick [5, 44, 45]. It has therefore been proposed that a signal from a peripheral target may initiate migration . To determine whether a signal is transported from the developing extraocular muscles back to the cell body to initiate migration, we utilized three techniques to eliminate interaction between the oculomotor nerve and the peripheral tissue, and examined oculomotor neuron migration.
To eliminate the possibility that peripheral signals derived from tissue other than the extraocular muscles initiate midline migration, we utilized a second in vitro technique where the midbrain was dissected away from the oculomotor nerve and peripheral tissues on E11.5. The midbrain, devoid of peripheral tissue, was cultured for 72 h in vitro, then oculomotor neurons were labeled. At the starting point on E11.5 (0 h), all oculomotor neurons reside on the edge of the floor plate with no Islet1/2-positive neurons in the floor plate (Fig. 11B). Following 72 h in vitro, Islet1/2-positive neurons were found throughout the floor plate, particularly in the posterior half of the nucleus where migration normally occurs in vivo (Fig. 11C). Therefore, isolated midbrain cultures provided sufficient cues to initiate and guide migration, and demonstrate that removal of the peripheral nerve does not appear to restrict oculomotor migration.
A caveat to the explant culture strategy was that the initial outgrowth of the nerve toward the eye occurs on E9.5. Therefore, a signal to activate migration could be transported back to the oculomotor nucleus prior to E11.5, although it must remain latent until overt migration begins on E13.5. To test the possibility that a signal is transported during the initial outgrowth of nIII, we examined a mutant mouse where the oculomotor nerve outgrowth into the periphery is absent. In CXCR4 mutant mice, axons projecting from the oculomotor nucleus wander dorsally within the neuroepithelium instead of toward the peripheral mesenchyme where the attractive CXCR4 ligand, SDF-1, is expressed . These mutants have an oculomotor nerve that aberrantly projects within the neuroepithelium, or is much smaller in size ; therefore, nIII would be unable to obtain or transport peripheral signals back to the nucleus to initiate migration. We first verified that the oculomotor nerve in CXCR4 mutant mice was missing, both by visual inspection during dissection and axon antibody labeling in whole mount embryos. We were unable to identify a peripheral oculomotor nerve emerging from the midbrain (data not shown), confirming the previous report, and suggesting that a signal from a peripheral intermediate target was unlikely to be transported back to the nucleus in CXCR4 mutant mice. As previously stated, we found Islet1/2-positive cell bodies in the floor plate of the midbrain in CXCR4 mutant mice (Fig. 9D), indicating that contralateral oculomotor migration can occur independent of nerve-derived signals obtained from outside of the neural tube. This suggests that signals to initiate migration are intrinsic to the oculomotor nucleus and nearby tissues within the ventral midbrain.
Biondi first observed a group of neurons that separate from the developing oculomotor column and migrate toward the opposing oculomotor nucleus . Later, through a series of Golgi labels, Puelles proposed that “diffusible factors” are likely responsible for guiding migration . Until our study, the identity of such factors remained unknown.
To initiate and guide the several steps involved in migration across the midline, we speculate that a combination of diffusible extrinsic signals and intrinsic factors are required. The migration of a subset of oculomotor neurons appears to involve multiple steps: 1) the superior rectus motor neurons receive extracellular or intracellular signals to initiate and extend a secondary leading process; 2) leading processes receive guidance signals to attract them toward the midline and eventually toward the contralateral nucleus; 3) the neuron cell bodies lose cell adhesion to the ipsilateral oculomotor nucleus and follow their secondary leading processes across the floor plate; 4) neurons must move to and integrate into the contralateral nucleus.
Here we show that Slit signaling is necessary to inhibit the initiation of oculomotor migration. Slits may function to gate migration, such that the suppression of Slit/Robo repulsion on E13.5 would allow a subset of oculomotor neurons to turn on migratory processes. Our data suggest a range of potential roles for Slit signaling in migratory OM neurons including; blocking extension of leading processes, promoting cell adhesion to ipsilateral nucleus, and blocking migration of the neuronal cell bodies across the floor plate.
Migrating oculomotor neurons pioneer an independent path across the midline via leading axon-like processes
Oculomotor neurons initially make a conventional axon projection that exits the CNS out toward their peripheral targets. However, then a subset undergo a remarkable transition to produce second axon-like fibers (leading processes) oriented toward and across the ventral midline, followed soon after by translocation of the neuronal cell bodies . We confirmed this time course in the mouse: all migrating oculomotor neurons possess a leading process that projects toward the midline, with a trajectory that is perpendicular to the radially-oriented ventricular cells in the floor plate on E13.5. The leading processes originate only from neurons in the caudal half of the nucleus, and can extend to reach the opposing oculomotor nucleus. These processes appear similar to leading processes found in migrating pontine neurons : leading processes are long, branching, and fasciculate with other processes.
Similar to pontine neurons, the guidance of oculomotor leading processes does not depend upon pre-existing glial structures  or the existing tecto- tegmental commissure located in the floor plate. A distinct gap between the existing tecto-tegmental commissure and migrating motor neurons can be found in every section through the oculomotor nucleus. Therefore, oculomotor leading projections must navigate using other environmental cues to reach their target, and suggests that they pioneer a path through a permissive corridor across the midline, independent of the tegmental commissure. The potential molecular and cellular substrates for the oculomotor commissure remain undefined.
The observation in the midline of both cell bodies and secondary leading processes suggested that contralateral cell bodies were not the result of their primary axons making midline crossing errors. The number of migrating motor neurons is surprisingly large, as in posterior sections they outnumber the remaining non-migratory neurons. Interestingly, a large degree of cell death in the ventromedial aspect of the chick oculomotor nucleus was previously noted . This suggests a large proportion of migratory neurons initiate migration but later succumb to cell death.
Slit signals prevent premature migration of oculomotor neurons
Guidance cues of the Slit family have emerged as repulsive regulators of midline crossing of axons and, more recently, in neuron cell bodies in both vertebrates and invertebrates reviewed by . In the ventral midbrain on E10.5, both Slit1 and 2 are expressed in the floor plate, and the Slit receptors, Robo1 and 2, expressed by the oculomotor neurons. Slit1 and 2 signals from the floor plate may actively repel the leading process, and hence oculomotor neurons, away from the floor plate via the Robo receptors. Consistent with this idea, we show significant premature outgrowth of leading processes in Slit1/2 or Robo1/2 mutants on E10.5. This is coupled with oculomotor cell migration across the floor plate in a subset of cells in the posterior region of the oculomotor nucleus. Previous research describes a similar role for Slit midline repulsion in the hindbrain where Slit1 and 2 are required to keep dorsally projecting motor neurons, but not ventrally projecting nerves, out of the floor plate . We also find that Slits are not required to guide the ventrally projecting nerve fibers, but are required to inhibit the growth of leading processes and migration of cell bodies across the floor plate. In chick, outgrowth of leading processes depends upon the actin-binding protein Drebrin . Therefore, Slit signaling may inhibit Drebrin activity to repress leading process outgrowth prior to E11.5. However, an interaction between Slit/Robo and Drebrin has yet to be described.
We found Slit2 and Slit3 transcripts in the oculomotor nucleus on E10.4 and E14.5. Slit2 and Slit3 also appear to be expressed by other motor neurons . In motor neurons, it is suggested that Slit2 and Slit3 act cell autonomously to modulate their own Robo receptor responsiveness , or are transported to the axon to promote nerve fasciculation . Our data from Isl1 mutant mice suggests that motor neuron-derived Slit2 does not participate in repelling oculomotor neurons from the floor plate on E11.5. However, Slit2 cell autonomous functions in the oculomotor nucleus remain to be clarified by future experiments with motor neuron-specific knockouts.
We find an increased number of cell bodies migrating across the midline in Robo mutants compared to Slit mutants on E10.5. This suggests that Robo may have additional ligands that mediate cell adhesion or inhibit cell migration. Signaling from the Robo receptor modulates cell adhesion by N-cadherin , and acts to either increase or decrease adhesion . However, cell adhesion in these systems is dependent upon the Slit ligand. Future research may give insight into how the Robo receptor differentiates between mediating cell adhesion and signaling repulsion.
Taken together, we propose that Slit/Robo signals keep leading processes away from the floor plate, and also possibly inhibit neural translocation within the processes. In a similar case, precerebellar neurons of the inferior olive (IO) respond to repulsive Slit signals from the floor plate. IO neurons normally project a leading process across the midline, but their cell body stops just prior to entering the floor plate in the hindbrain. In Robo1/2 mutant mice, IO cell bodies migrate across the floor plate, indicating that Slits are acting as midline repellents via the Robo1/2 receptors [53, 54]. The leading processes in the IO are not normally repelled from the midline, indicating that these two actions—leading process guidance and neural translocation— are independent of each other, with neural translocation being susceptible to repellent Slit signals in IO neurons. However, in the oculomotor system, both leading process outgrowth and neural translocation appear equally responsive to repellent Slit signals.
Migration into the floor plate on E10.5 in Slit and Robo mutants is seen in only a subset of oculomotor neurons. These neurons emanate from the caudal half of the ipsilateral nucleus and are competent to cross the floor plate to reside in the contralateral nucleus. Wild type migration of superior rectus motor neurons occurs similarly, starting with a leading process projecting from the caudal half of the ipsilateral nucleus toward the contralateral nucleus, neuronal migration, and finally integration into the contralateral nucleus. The caudal location and size of the commissure suggests that oculomotor neurons that migrate on E10.5 in Slit and Robo mutants are superior rectus motor neurons that migrate prematurely. Premature projection of retinal ganglion cell axons also occurs in Slit1 and 2 mutant mice . Premature migration suggests that superior rectus motor neurons are responsive to Slit repulsion at the floor plate prior to E13.5 and that Slit/Robo repulsion keeps the motor neurons in their initial ipsilateral positions until this later time point. However, a method to specifically label superior rectus motor neurons is not currently available, so we were not able to distinguish which sub-populations of oculomotor neurons were recruited to cross the midline in Slit or Robo mutants.
Factors that regulate the timing of Slit repulsion of motor neurons
We suggest that Slits initially repel superior rectus motor neurons on E10.5. Later, on E13.5, this subset must become insensitive to Slit repulsion to enter the floor plate. This could occur by blocking the repulsive activity of floor plate-derived Slit, or increasing oculomotor attraction to the floor plate. Turning off the Slit signal could occur by a general decrease of Robo on the surface of migrating neurons. For example, in Drosophila, the cytoplasmic protein Commissureless acts to reduce Robo receptors at the cell surface thereby inhibiting repulsive Slit signaling [55, 56]. However, Commissureless homologues have not yet been identified in vertebrates, so it remains unclear whether an analogous mechanism might be involved with vertebrate Robos.
Activating receptors that compete with, or block Slit/ Robo signaling, may be another strategy to release neurons to migrate into Slit positive territory. We tested the possibilities that receptors Robo3 or CXCR4 may interfere with Slit signaling allow for Robo-expressing oculomotor neurons to migrate into zones of high Slit expression. Although some evidence suggested that oculomotor neurons express Robo3 and CXCR4, we found that these receptors are not required for midline crossing. We also found that Netrin1 expressed by the ventral floor plate is not required for migrating oculomotor neurons. This may indicate that other attractive proteins function to attract oculomotor neurons either in concert with Netrin1 or independently. Other signals that block Slit/Robo signaling include NPN1. NPN1 interacts with Robo1 to reduce the repulsive effects of Semaphorin3A in cortical interneurons . In chick, class 3 Semaphorins are expressed in the oculomotor nucleus as well as in the ventral floor plate, while NPN1 is found in migrating oculomotor neurons . Therefore, NPN1 receptors may interact with Robo receptors on superior rectus motor neurons to allow for migration. However, we were unable to detect NPN1 antibody labeling in the oculomotor nucleus (data not shown). Future work may investigate the contribution of Semaphorin and Neuropilin signaling in mediating oculomotor migration.
Oculomotor migration does not rely on signals from target tissues
The coincidental timing of superior rectus extraocular muscle innervation and motor neuron migration suggests that a signal obtained from the muscle or intermediate target may be transported from peripheral tissue to the cell body to initiate migration . However, we found that oculomotor neurons migrate in the absence of extraocular muscles, peripheral tissue, and the peripheral nIII itself. These results agree with previous findings that superior rectus motor neurons migrate in Phox2b knock-in mice that lack the oculomotor nerve . Signals from within the neural tube therefore provide the signal to initiate migration. For example, the idea that the tecto-tegmental commissural axons traveling from the dorsal midbrain toward and across the floor plate may initiate oculomotor migration  was tested by ablation of the commissural axons during chick development. However, oculomotor neuron migration was not altered . Likewise, it was suggested that dopamine generated by midline cells that later generate the substantia nigra and ventral tegmentum may attract oculomotor neurons . However, there is no evidence thus far that monoamines have chemoattractant properties.
Therefore, other signals from within the neural tube, likely in conjunction with intrinsic properties such as cell surface receptors, specific to superior rectus motor neurons, may initiate the outgrowth of the leading projection and subsequent neural migration.
Migration of superior rectus motor neurons across the ventral midline to the contralateral oculomotor nucleus likely requires tight regulation by a number of extrinsic cues and intrinsic responses. Mice lacking Slit1 and 2 or Robo1 and 2 have motor neurons that migrate to the contralateral nucleus on E10.5, three days prior to the normal migration on E13.5. This suggests active Slit/Robo repulsion from the midline is required, at least at early stages, to maintain the ipsilateral position of oculomotor neurons. Neurons that migrate on E10.5 in Slit1/2 or Robo1/2 mutants are positioned in the caudal half of the nucleus, suggesting these neurons are superior rectus motor neurons migrating prematurely. Migration across the midline at E10.5 requires competency to interpret midline attraction, as well as attraction toward the contralateral nucleus. The classic midline attractant Netrin1, or its receptor DCC, does not appear to be required for successful superior rectus motor neuron migration. However, other attractive signals working independently of, or redundant to, Netrin1 are available as early as E10.5 to promote midline crossing in Slit and Robo mutant mice. Furthermore, the ability to respond to and initiate migration across the midline is intrinsic to the neural tube, and does not require signals obtained from the peripheral tissue.
Deleted in cortical cancer
Long Chain Dialkylcarbocyanine, Lipophilic Dyes
Medial longitudinal fasciculus
Messenger ribonucleic acid
Cranial nerve three
Phosphate buffered saline
Phosphate buffered saline with triton detergent
Sonic hedge hog
We thank Marc Tessier-Lavigne (Rockefeller) for Slit1;2;3 and Robo1 and 2 mutant mouse lines, and in situ probes. We thank Marc Tessier-Lavigne and Fred Charron (IRCM, Montreal) for Netrin1 and DCC mutant mouse lines, John Rubenstein and Daniel Vogt (UCSF) for CXCR4 embryos, Alain Chedotal (INSERM, Paris) for providing Robo3 mutant mouse embryos and Robo3 expression data, Philip Gage and Lisheng Chen (U Michigan) for Pitx2 embryos, and Elke Stein (Yale) for providing Robo1 and 2 antibodies. We thank Andrew Roesener for providing images of emigrating neurons. We thank Katie Weller and Christine Schlemmer for assistance with experiments, insights and helpful comments.
Funding sources included NIH grants NS054740, NS077169, and EY025205 to GSM, and NRF of Korea (NRF-2013R1A1A2058548), Integrative Aging Research Center of Gwangju Institute of Science and Technology and Korea Health Industry Development Institute (HI14C3484) to MRS. Further support for core facilities at the University of Nevada was provided by NIH COBREs RR024210, GM103650, GM103554, and the Nevada INBRE 8 P20 GM103440-11, P20 GM103554, with use of the Nevada Genomics Center supported by INBRE (NCRR) P20 RR-016464. The funding bodies had no roles in the study, collection, analysis, interpretation of data, nor in writing the manuscript.
Availability of data and materials
The materials used in this study will be made available on request.
BB and GSM developed the project. TF, ER, and BB performed in situ hybridization, FST performed explants, and BB performed nerve tracing, antibody labeling and analysis. MKK performed Robo antibody labeling. KTK and MRS performed the analysis of Islet mutant embryos. BB and GSM wrote the paper. All authors read and approved the manuscript.
The authors declare they have no competing interests.
Animal experiments were approved by the UNR IACUC, following NIH guidelines, with the approved protocol #2015-00435.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Granet DB, Khayali S. Amblyopia and strabismus. Pediatr Ann. 2011;40:89–94.View ArticlePubMedGoogle Scholar
- Puelles L, Privat A. Do oculomotor neuroblasts migrate across the midline in the retal rat brain? Anat Embryol (Berl). 1977;150:187–206.View ArticleGoogle Scholar
- Puelles L. A Golgi-study of oculomotor neuroblasts migrating across the midline in chick embryos. Anat Embryol (Berl). 1978;152:205–15.View ArticleGoogle Scholar
- Hogg ID. Observations of the development of the nucleus of Edinger-Westphal in man and the albino rat. J Comp Neurol. 1966;126:567–84.PubMedGoogle Scholar
- Puelles-Lopez L, Malagon-Cobos F, Genis-Galvez JM. The Migration of Oculomotor Neuroblasts Across the Midline in Chick Embryo. Exp Neurol. 1975;47:459–69.View ArticleGoogle Scholar
- Dun XP, Bandeira de Lima T, Allen J, Geraldo S, Gordon-Weeks P, Chilton JK. Drebrin controls neuronal migration through the formation and alignment of the leading process. Mol Cell Neurosci. 2012;49:341–50.View ArticlePubMedPubMed CentralGoogle Scholar
- Cheng L, Desai J, Miranda CJ, Duncan JS, Qiu W, Nugent AA, Kolpak AL, Wu CC, Drokhlyansky E, Delisle MM, et al. Human CFEOM1 mutations attenuate KIF21A autoinhibition and cause oculomotor axon stalling. Neuron. 2014;82:334–49.View ArticlePubMedPubMed CentralGoogle Scholar
- Naujoks-Manteuffel C, Sonntag R, Fritzsch B. Development of the amphibian oculomotor complex: evidences for migration of oculomotor motorneurons across the midline. Anat Embryol. 1991;183:545–52.View ArticlePubMedGoogle Scholar
- Colamarino SA, Tessier-Lavigne M. The role of the floor plate in axon guidance. Annu Rev Neurosci. 1995;18:497–529.View ArticlePubMedGoogle Scholar
- Dickson BJ, Gilestro GF. Regulation of commissural axon pathfinding by slit and its Robo receptors. Annu Rev Cell Dev Biol. 2006;22:651–75.View ArticlePubMedGoogle Scholar
- Wong K, Park H, Wu J, Rao Y. Slit proteins: molecular guidance cues for cells ranging from neurons to leukocytes. Curr Opin Genet Dev. 2002;12:583–91.View ArticlePubMedGoogle Scholar
- Plump AS, Erskine L, Sabatier C, Brose K, Epstein CJ, Goodman CS, Mason CA, Tessier-Lavigne M. Slit1 and Slit2 cooperate to prevent premature midline crossing of retinal axons in the mouse visual system. Neuron. 2002;33:219–32.View ArticlePubMedGoogle Scholar
- Farmer WT, Altick AL, Nural HF, Dugan JP, Kidd T, Charron F, Mastick GS. Pioneer longitudinal axons navigate using floor plate and Slit/Robo signals. Development. 2008;135:3643–53.View ArticlePubMedPubMed CentralGoogle Scholar
- Long H, Sabatier C, Ma L, Plump A, Yuan W, Ornitz DM, Tamada A, Murakami F, Goodman CS, Tessier-Lavigne M. Conserved roles for Slit and Robo proteins in midline commissural axon guidance. Neuron. 2004;42:213–23.View ArticlePubMedGoogle Scholar
- Sabatier C, Plump AS, Le M, Brose K, Tamada A, Murakami F, Lee EY, Tessier-Lavigne M. The divergent Robo family protein rig-1/Robo3 is a negative regulator of slit responsiveness required for midline crossing by commissural axons. Cell. 2004;117:157–69.View ArticlePubMedGoogle Scholar
- Zelina P, Blockus H, Zagar Y, Peres A, Friocourt F, Wu Z, Rama N, Fouquet C, Hohenester E, Tessier-Lavigne M, et al. Signaling switch of the axon guidance receptor Robo3 during vertebrate evolution. Neuron. 2014;84:1258–72.View ArticlePubMedGoogle Scholar
- Bloch-Gallego E, Ezan F, Tessier-Lavigne M, Sotelo C. Floor plate and netrin-1 are involved in the migration and survival of inferior olivary neurons. J Neurosci. 1999;19:4407–20.PubMedGoogle Scholar
- Causeret F, Danne F, Ezan F, Sotelo C, Bloch-Gallego E. Slit antagonizes netrin-1 attractive effects during the migration of inferior olivary neurons. Dev Biol. 2002;246:429–40.View ArticlePubMedGoogle Scholar
- Kennedy TE, Serafini T, de la Torre JR, Tessier-Lavigne M. Netrins are diffusible chemotropic factors for commissural axons in the embryonic spinal cord. Cell. 1994;78:425–35.View ArticlePubMedGoogle Scholar
- Keino-Masu K, Masu M, Hinck L, Leonardo ED, Chan SS, Culotti JG, Tessier-Lavigne M. Deleted in Colorectal Cancer (DCC) encodes a netrin receptor. Cell. 1996;87:175–85.View ArticlePubMedGoogle Scholar
- Shoja-Taheri F, DeMarco A, Mastick GS. Netrin1-DCC-Mediated Attraction Guides Post-Crossing Commissural Axons in the Hindbrain. J Neurosci. 2015;35:11707–18.View ArticlePubMedPubMed CentralGoogle Scholar
- Varela-Echavarría A, Tucker A, Püschel AW, Guthrie S. Motor axon subpopulations respond differentially to the chemorepellents netrin-1 and semaphorin D. Neuron. 1997;18:193–207.View ArticlePubMedGoogle Scholar
- Hammond R, Vivancos V, Naeem A, Chilton J, Mambetisaeva E, Andrews W, Sundaresan V, Guthrie S. Slit-mediated repulsion is a key regulator of motor axon pathfinding in the hindbrain. Development. 2005;132:4483–95.View ArticlePubMedGoogle Scholar
- Murray A, Naeem A, Barnes SH, Drescher U, Guthrie S. Slit and Netrin-1 guide cranial motor axon pathfinding via Rho-kinase, myosin light chain kinase and myosin II. Neural Dev. 2010;5:15.View ArticleGoogle Scholar
- Lin L, Rao Y, Isacson O. Netrin-1 and slit-2 regulate and direct neurite growth of ventral midbrain dopaminergic neurons. Mol Cell Neurosci. 2005;28:547–55.View ArticlePubMedGoogle Scholar
- Dugan JP, Stratton A, Riley HP, Farmer WT, Mastick GS. Midbrain dopaminergic axons are guided longitudinally through the diencephalon by Slit/Robo signals. Mol Cell Neurosci. 2011;46:347–56.View ArticlePubMedGoogle Scholar
- Fazeli A, Dickinson SL, Hermiston ML, Tighe RV, Steen RG, Small CG, Stoeckli ET, Keino-Masu K, Masu M, Rayburn H, et al. Phenotype of mice lacking functional Deleted in colorectal cancer (Dcc) gene. Nature. 1997;386:796–804.View ArticlePubMedGoogle Scholar
- Serafini T, Colamarino SA, Leonardo ED, Wang H, Beddington R, Skarnes WC, Tessier-Lavigne M. Netrin-1 is required for commissural axon guidance in the developing vertebrate nervous system. Cell. 1996;87:1001–14.View ArticlePubMedGoogle Scholar
- Grieshammer U, Le M, Plump AS, Wang F, Tessier-Lavigne M, Martin GR. SLIT2-mediated ROBO2 signaling restricts kidney induction to a single site. Dev Cell. 2004;6:709–17.View ArticlePubMedGoogle Scholar
- Mastick GS, Davis NM, Andrew GL, Easter Jr SS. Pax-6 functions in boundary formation and axon guidance in the embryonic mouse forebrain. Development 1997, 124:1985-1997
- Kim M, Roesener AP, Mendonca PR, Mastick GS. Robo1 and Robo2 have distinct roles in pioneer longitudinal axon guidance. Dev Biol. 2011;358:181–8.View ArticlePubMedPubMed CentralGoogle Scholar
- Ericson J, Thor S, Edlund T, Jessell TM, Yamada T. Early stages of motor neuron differentiation revealed by expression of homeobox gene Islet-1. Science. 1992;256:1555–60.View ArticlePubMedGoogle Scholar
- Carpenter F. The development of the oculomotor nerve, the ciliary ganglion, and the abducent nerve in the chick. Cambridge: Bulletin of the museum of comparative zoology; 1906.View ArticleGoogle Scholar
- Chedotal A, Richards LJ. Wiring the brain: the biology of neuronal guidance. Cold Spring Harb Perspect Biol. 2010;2:a001917.View ArticlePubMedPubMed CentralGoogle Scholar
- Hasan KB, Agarwala S, Ragsdale CW. PHOX2A regulation of oculomotor complex nucleogenesis. Development. 2010;137:1205–13.View ArticlePubMedPubMed CentralGoogle Scholar
- Brose K, Bland KS, Wang KH, Arnott D, Henzel W, Goodman CS, Tessier-Lavigne M, Kidd T. Slit proteins bind Robo receptors and have an evolutionarily conserved role in repulsive axon guidance. Cell. 1999;96:795–806.View ArticlePubMedGoogle Scholar
- Ricano-Cornejo I, Altick AL, Garcia-Pena CM, Nural HF, Echevarria D, Miquelajauregui A, Mastick GS, Varela-Echavarria A. Slit-Robo signals regulate pioneer axon pathfinding of the tract of the postoptic commissure in the mammalian forebrain. J Neurosci Res. 2011;89:1531–41.View ArticlePubMedPubMed CentralGoogle Scholar
- Lee H, Kim M, Kim N, Macfarlan T, Pfaff SL, Mastick GS, Song MR. Slit and Semaphorin signaling governed by Islet transcription factors positions motor neuron somata within the neural tube. Exp Neurol. 2015;269:17–27.View ArticlePubMedPubMed CentralGoogle Scholar
- Chen Z, Gore BB, Long H, Ma L, Tessier-Lavigne M. Alternative splicing of the Robo3 axon guidance receptor governs the midline switch from attraction to repulsion. Neuron. 2008;58:325–32.View ArticlePubMedGoogle Scholar
- Chalasani SH, Sabelko KA, Sunshine MJ, Littman DR, Raper JA. A chemokine, SDF-1, reduces the effectiveness of multiple axonal repellents and is required for normal axon pathfinding. J Neurosci. 2003;23:1360–71.PubMedGoogle Scholar
- Chalasani SH, Sabol A, Xu H, Gyda MA, Rasband K, Granato M, Chien CB, Raper JA. Stromal cell-derived factor-1 antagonizes slit/robo signaling in vivo. J Neurosci. 2007;27:973–80.View ArticlePubMedGoogle Scholar
- Lerner O, Davenport D, Patel P, Psatha M, Lieberam I, Guthrie S. Stromal cell-derived factor-1 and hepatocyte growth factor guide axon projections to the extraocular muscles. Dev Neurobiol. 2010;70:549–64.View ArticlePubMedGoogle Scholar
- Yee KT, Simon HH, Tessier-Lavigne M, O'Leary DM. Extension of long leading processes and neuronal migration in the mammalian brain directed by the chemoattractant netrin-1. Neuron. 1999;24:607–22.View ArticlePubMedGoogle Scholar
- Chilton JK, Guthrie S. Development of oculomotor axon projections in the chick embryo. J Comp Neurol. 2004;472:308–17.View ArticlePubMedGoogle Scholar
- Fritzsch B, Nichols DH, Echelard Y, McMahon AP. Development of midbrain and anterior hindbrain ocular motoneurons in normal and Wnt-1 knockout mice. J Neurobiol. 1995;27:457–69.View ArticlePubMedGoogle Scholar
- Zacharias AL, Lewandoski M, Rudnicki MA, Gage PJ. Pitx2 is an upstream activator of extraocular myogenesis and survival. Dev Biol. 2011;349:395–405.View ArticlePubMedGoogle Scholar
- Biondi G. Osservaxioni sullo sviluppo e sulla struttura dei nuclei d'origine dei nervi oculomotore e trocleare nel pollo. Rivista Italiana Neuropathia. 1910;3:302–403.Google Scholar
- Steljes TP, Kinoshita Y, Wheeler EF, Oppenheim RW, von Bartheld CS. Neurotrophic factor regulation of developing avian oculomotor neurons: differential effects of BDNF and GDNF. J Neurobiol. 1999;41:295–315.View ArticlePubMedGoogle Scholar
- Ypsilanti AR, Zagar Y, Chedotal A. Moving away from the midline: new developments for Slit and Robo. Development. 2010;137:1939–52.View ArticlePubMedGoogle Scholar
- Jaworski A, Tessier-Lavigne M. Autocrine/juxtaparacrine regulation of axon fasciculation by Slit-Robo signaling. Nat Neurosci. 2012;15:367–9.View ArticlePubMedGoogle Scholar
- Rhee J, Buchan T, Zukerberg L, Lilien J, Balsamo J. Cables links Robo-bound Abl kinase to N-cadherin-bound beta-catenin to mediate Slit-induced modulation of adhesion and transcription. Nat Cell Biol. 2007;9:883–92.View ArticlePubMedGoogle Scholar
- Shiau CE, Bronner-Fraser M. N-cadherin acts in concert with Slit1-Robo2 signaling in regulating aggregation of placode-derived cranial sensory neurons. Development. 2009;136:4155–64.View ArticlePubMedPubMed CentralGoogle Scholar
- Marillat V, Sabatier C, Failli V, Matsunaga E, Sotelo C, Tessier-Lavigne M, Chedotal A. The slit receptor Rig-1/Robo3 controls midline crossing by hindbrain precerebellar neurons and axons. Neuron. 2004;43:69–79.View ArticlePubMedGoogle Scholar
- Di Meglio T, Nguyen-Ba-Charvet KT, Tessier-Lavigne M, Sotelo C, Chedotal A. Molecular mechanisms controlling midline crossing by precerebellar neurons. J Neurosci. 2008;28:6285–94.View ArticlePubMedGoogle Scholar
- Kidd T, Brose K, Mitchell KJ, Fetter RD, Tessier-Lavigne M, Goodman CS, Tear G. Roundabout controls axon crossing of the CNS midline and defines a novel subfamily of evolutionarily conserved guidance receptors. Cell. 1998;92:205–15.View ArticlePubMedGoogle Scholar
- Tear G, Harris R, Sutaria S, Kilomanski K, Goodman CS, Seeger MA. commissureless controls growth cone guidance across the CNS midline in Drosophila and encodes a novel membrane protein. Neuron. 1996;16:501–14.View ArticlePubMedGoogle Scholar
- Hernandez-Miranda LR, Cariboni A, Faux C, Ruhrberg C, Cho JH, Cloutier JF, Eickholt BJ, Parnavelas JG, Andrews WD. Robo1 regulates semaphorin signaling to guide the migration of cortical interneurons through the ventral forebrain. J Neurosci. 2011;31:6174–87.View ArticlePubMedPubMed CentralGoogle Scholar
- Coppola E, Pattyn A, Guthrie SC, Goridis C, Studer M. Reciprocal gene replacements reveal unique functions for Phox2 genes during neural differentiation. EMBO J. 2005;24:4392–403.View ArticlePubMedPubMed CentralGoogle Scholar
- Goldberg S. Studies on the mechanics of development of the visual pathways in the chick embryo. Dev Biol. 1974;36:24–43.View ArticlePubMedGoogle Scholar
- Heaton MB, Moody SA, Coultas PL. Oculomotor neuroblast migration in the chick embryo in the absence of tecto-tegmental fibers. Dev Biol. 1979;68:304–10.View ArticlePubMedGoogle Scholar