Dual function of Slit2 in repulsion and enhanced migration of trunk, but not vagal, neural crest cells

Abstract

Neural crest precursors to the autonomic nervous system form different derivatives depending upon their axial level of origin; for example, vagal, but not trunk, neural crest cells form the enteric ganglia of the gut. Here, we show that Slit2 is expressed at the entrance of the gut, which is selectively invaded by vagal, but not trunk, neural crest. Accordingly, only trunk neural crest cells express Robo receptors. In vivo and in vitro experiments demonstrate that trunk, not vagal, crest cells avoid cells or cell membranes expressing Slit2, thereby contributing to the differential ability of neural crest populations to invade and innervate the gut. Conversely, exposure to soluble Slit2 significantly increases the distance traversed by trunk neural crest cells. These results suggest that Slit2 can act bifunctionally, both repulsing and stimulating the motility of trunk neural crest cells.

Figure 1.

Expression of Slits during neural crest migration. In situ hybridization of Slit2, 1, and 3 demonstrates that Slits are expressed in the mesenchyme at the entry to the gut as well as in the neural tube and somites. (a–c) Whole mounts of stage 17 chicken embryos. (a) Whole mount in situ hybridization with a Slit2 probe reveals that it is expressed in the dorsal neural tube (black arrow), ventral neural tube (white arrow), and gut mesenchyme (black arrowhead). (b and c) Whole mount in situ hybridization with a Slit1 probe reveals that it is expressed in the dorsal neural tube, dorsomedial dermomyotome (red arrow), and gut mesenchyme. (b) A higher magnification of c. (d) Transverse section in situ hybridization through the lumbar level of a stage 17 chick embryo shows Slit2 in the floor plate and developing motor neurons (white arrow), the roof plate of the neural tube (black arrow), and the mesenchyme immediately dorsal to the gut (black arrowhead); the Slit1 pattern (not depicted) looked identical. (e) A similar expression pattern for Slit2 and Slit1 (not depicted) was observed at the hindlimb level of a stage 19 embryo. (f) Slit3 in a stage 16 embryo appeared similar to the other Slits, except that staining was reduced or absent in the floor plate and dorsal neural tube.

Figure 2.

Expression of Robos during neural crest migration. Whole mount in situ hybridization with Robo1 and Robo2 probe reveals that the receptor is expressed in the trunk neural tube, on migrating trunk neural crest cells within the somites, and in the dermomyotome, but is not expressed by vagal neural crest cells. (a) In a stage 15 chicken embryo, there is no staining for Robo2 at vagal levels (bracket) though there is staining at truncal levels. (b and c) At stage 18 during the peak of neural crest migration, Robo1 (b) and Robo2 (c) are strongly expressed in neural crest cells at the trunk levels (black arrows) but not in the vagal region (brackets). (d) Robo2 labeling in the head of a stage 15 embryo shows staining on the dorsal half of the otic vesicle (ov), trigeminal placode, and dorsal neurons in the mesencephalon; the posterior part of the second/hyoid branchial arch is also positive for Robo2 (arrowhead). (e–g) At stage 20, when neural crest cells are condensing to form the dorsal root ganglia (arrows), Robo1 (e) and Robo2 (g) are strongly expressed in trunk, but not vagal (brackets), regions. (f) The same stage embryo labeled with HNK-1 shows neural crest cells migrating and beginning to condense into dorsal root ganglia at both vagal and trunk levels.

Figure 3.

Expression of Robos on migrating trunk but not vagal neural crest. Section in situ hybridization with Robo1 and Robo2 probes reveals that the receptor is expressed in the trunk neural tube and on migrating trunk neural crest cells within the somites but is not expressed by vagal neural crest cells. All embryos were stage 18. Left panels show in situ signal, and right panels show the same section stained with HNK-1 antibody to recognize neural crest cells. (a and b) Robo1 is strongly expressed in trunk neural crest cells (arrows) and motor neuron precursors in the ventral neural tube (NT). (c and d) Robo2 is also strongly expressed in migrating trunk neural crest cells and the neural tube, except for the ventral-most side. (e–h) At vagal levels, Robo1 (e and f) and Robo2 (g and h) are expressed in the dermomyotome and neural tube but not in the migrating vagal neural crest cells.

Figure 4.

Effects of Slit2-expressing cells on neural crest migration in vivo. Cells expressing Slit2 or control HEK cells were labeled with the lipophilic dye DiI and implanted onto vagal and/or trunk neural crest migratory pathways. Left panel shows flattened confocal Z-series of whole mounts of embryos stained with the HNK-1 antibody (green) to recognize neural crest cells, and the right panel shows both the neural crest and the injected cells (red) at both vagal and trunk levels. Embryos were analyzed one day after injection. (a–d) At vagal levels, neural crest cells intermixed with both control and Slit2-expressing cells. (g–l) In contrast, at trunk levels, neural crest cells overlapped with control cells (g and h) but appeared to stop (white arrowhead) some distance away from Slit2 cells (i–l). Notice also how trunk neural crest cells circumvent Slit2-expressing cells (white arrow in j). (e and f) Sections through embryos injected with Slit2 show that vagal neural crest freely intermix with Slit2 cells, whereas trunk neural crest cells avoid (white arrowhead) Slit2 cells.

Figure 5.

Trunk, not vagal, neural crest cells are repelled by Slit2- expressing cells in vitro. (a–d) Trunk and vagal neural crest cells (green) were grown apposed to live control HEK cells (a and c) or Slit2-expressing cells (b and d) (blue DAPI label). Only when trunk neural crest cells (b) were grown with Slit2-expressing cells was there a sharp border (white arrowhead) formed between the two populations. (e–h) A similar experiment performed with dead control or Slit2-expressing cells again shows a border between trunk neural crest cells and the Slit cell ghosts (white arrowhead in f), demonstrating that the repellent activity is membrane bound.

Figure 6.

Neural crest cells migrate further in the presence of Slit2 CM. (a) A neural tube explanted in the presence of CM from control cells shows that migrating neural crest cells have moved several cell diameters away from the neural tube after 18 h in culture. (b) A similar neural tube cultured in medium conditioned by Slit2-expressing cells had neural crest cells that had migrated significantly further in 18 h.

Figure 7.

Slit2 enhances trunk neural crest motility in a wound assay. Trunk neural tubes were cultured overnight on fibronectin. After one day, media was changed to one conditioned by control or Slit2-secreting cells. A wound of one to two cells width was made with a fine pipette. After 2 h, the percent of wounds with cells crossing and sealing the gap was determined. (a, unsealed) Image of a neural crest culture fixed immediately after wounding and stained with the HNK-1 antibody. (a, sealed) Image of a similar culture fixed 4 h after wounding showing that many neural crest cells have sealed the gap by this time point. (b) Primed neural crest cultures were incubated for 2 h before performing the wound with media conditioned for 5 d by control HEK cells or Slit2-secreting cells. Nonprimed corresponds to neural crest cells that were not preexposed to Slit2 in the media before the wound. Data correspond to one representative experiment out of eight. (c) Slit2 enhances wound healing of trunk, not vagal, neural crest, and this effect can be reversed by soluble Robo. The enhanced migration of trunk neural crest by Slit2 was significantly reduced by the presence of RoboN in the media. Neural crest cultures were primed for 2 h before performing the wound with media conditioned for 5 d by control HEK cells, Slit2-secreting cells, or a 1:1 combination of RoboN and Slit2 media. After 2 h of culture, the percent of wounds with cells crossing and sealing the gap was determined. Data correspond to one representative experiment of six.

Figure 7.

Slit2 enhances trunk neural crest motility in a wound assay. Trunk neural tubes were cultured overnight on fibronectin. After one day, media was changed to one conditioned by control or Slit2-secreting cells. A wound of one to two cells width was made with a fine pipette. After 2 h, the percent of wounds with cells crossing and sealing the gap was determined. (a, unsealed) Image of a neural crest culture fixed immediately after wounding and stained with the HNK-1 antibody. (a, sealed) Image of a similar culture fixed 4 h after wounding showing that many neural crest cells have sealed the gap by this time point. (b) Primed neural crest cultures were incubated for 2 h before performing the wound with media conditioned for 5 d by control HEK cells or Slit2-secreting cells. Nonprimed corresponds to neural crest cells that were not preexposed to Slit2 in the media before the wound. Data correspond to one representative experiment out of eight. (c) Slit2 enhances wound healing of trunk, not vagal, neural crest, and this effect can be reversed by soluble Robo. The enhanced migration of trunk neural crest by Slit2 was significantly reduced by the presence of RoboN in the media. Neural crest cultures were primed for 2 h before performing the wound with media conditioned for 5 d by control HEK cells, Slit2-secreting cells, or a 1:1 combination of RoboN and Slit2 media. After 2 h of culture, the percent of wounds with cells crossing and sealing the gap was determined. Data correspond to one representative experiment of six.

Figure 7.

Slit2 enhances trunk neural crest motility in a wound assay. Trunk neural tubes were cultured overnight on fibronectin. After one day, media was changed to one conditioned by control or Slit2-secreting cells. A wound of one to two cells width was made with a fine pipette. After 2 h, the percent of wounds with cells crossing and sealing the gap was determined. (a, unsealed) Image of a neural crest culture fixed immediately after wounding and stained with the HNK-1 antibody. (a, sealed) Image of a similar culture fixed 4 h after wounding showing that many neural crest cells have sealed the gap by this time point. (b) Primed neural crest cultures were incubated for 2 h before performing the wound with media conditioned for 5 d by control HEK cells or Slit2-secreting cells. Nonprimed corresponds to neural crest cells that were not preexposed to Slit2 in the media before the wound. Data correspond to one representative experiment out of eight. (c) Slit2 enhances wound healing of trunk, not vagal, neural crest, and this effect can be reversed by soluble Robo. The enhanced migration of trunk neural crest by Slit2 was significantly reduced by the presence of RoboN in the media. Neural crest cultures were primed for 2 h before performing the wound with media conditioned for 5 d by control HEK cells, Slit2-secreting cells, or a 1:1 combination of RoboN and Slit2 media. After 2 h of culture, the percent of wounds with cells crossing and sealing the gap was determined. Data correspond to one representative experiment of six.

Figure 8.

Movie stills from time-lapse video microscopy. Trunk neural crest cells exposed to Slit2 migrate further and have a longer total path length than those exposed to control medium. Trunk neural crest cells were labeled with Calcein AM (Molecular Probes) and washed before exposure to control or Slit2 CM. Cultures were time lapsed for ∼2.5 h under a confocal microscope. Images represent stills from a movie taken at the indicated times. Two cells (red) in each movie were manually traced to follow their movements. Their final path length is indicated in yellow in the last frame.

Figure 9.

Slit2 enhances neural crest cell migration. Trunk neural crest cells exposed to Slit2 migrate for longer distances compared with control exposed neural crest cells. Trunk and vagal neural tubes were cultured overnight on fibronectin. After one day, media was changed to one conditioned by control or Slit2-secreting cells 1 h before video microscopy in a confocal microscope for 3 h. (a) The total path length was determined and normalized to a 2.5-h time period and binned in groups of 100-μm distances traveled. (b) Total path length (total distance traveled, including the various turns made by the cells) of neural crest cells was plotted as cumulative percentages of the distance traveled. (c) Net path length (net distance from starting point) of neural crest cells was plotted as cumulative percentages of the distance traveled.