August 1, 1997;
The EphA4 and EphB1 receptor tyrosine kinases and ephrin-B2 ligand regulate targeted migration of branchial neural crest cells.
BACKGROUND: During vertebrate head
development, neural crest cells migrate from hindbrain
segments to specific branchial arches, where they differentiate into distinct patterns of skeletal structures. The rostrocaudal identity of branchial neural crest cells appears to be specified prior to migration, so it is important that they are targeted to the correct destination. In Xenopus embryos, branchial neural crest cells segregate into four streams that are adjacent during early stages of migration. It is not known what restricts the intermingling of these migrating cell populations and targets them to specific branchial arches. Here, we investigated the role of Eph receptors and ephrins-mediators of cell-contact-dependent interactions that have been implicated in neuronal pathfinding-in this targeted migration.
RESULTS: Xenopus EphA4
are expressed in migrating neural crest cells and mesoderm
of the third arch, and third plus fourth arches, respectively. The ephrin-B2 ligand, which interacts with these receptors, is expressed in the adjacent second arch neural crest and mesoderm
. Using truncated receptors, we show that the inhibition of EphA4
function leads to abnormal migration of third arch neural crest cells into second and fourth arch territories. Furthermore, ectopic activation of these receptors by overexpression of ephrin-B2 leads to scattering of third arch neural crest cells into adjacent regions. Similar disruptions occur when the expression of ephrin-B2 or truncated receptors is targeted to the neural crest.
CONCLUSIONS: These data indicate that the complementary expression of EphA4
receptors and ephrin-B2 is involved in restricting the intermingling of third and second arch neural crest and in targeting third arch neural crest to the correct destination. Together with previous work showing that Eph receptors and ligands mediate neuronal growth cone repulsion, our findings suggest that similar mechanisms are used for neural crest and axon
[+] show captions
Figure 1. Restricted migration of branchial neural crest cells in Xenopus. (a) The initial segmentation and (b,c) migration of streams of neural crest into specific branchial arches (1–4). The third and fourth arch neural crest streams segment during migration. Adapted from . (d) Expression pattern of AP-2 plus Krox-20 in a stage 25 Xenopus embryo revealed by whole-mount in situ hybridisation. AP-2 is expressed in all branchial neural crest (n1–n4: neural crest streams migrating into branchial arches 1–4), whereas Krox-20 expression is restricted to third arch neural crest (see also Figure 2f), which therefore is stained more intensely. (e) Section through dorsal region of migrating neural crest after detection of AP-2 plus Krox-20 expression. The neural crest destined for the second, third and fourth arches are adjacent. (f) Section through a more ventral region of the embryo shown in (d). After migration deep into the branchial arches, the streams of second, third and fourth arch crest are now separated by the pharyngeal pouches of the endoderm.
Figure 2. Expression of EphA4 and EphB1 in the branchial region. Whole-mount in situ hybridisation was carried out to detect EphA4 and EphB1 expression during branchial neural crest migration. (a–e) Expression pattern of EphA4. (a) A stage 19 embryo shown in dorsal–rostral view. Note the ventral domain of EphA4 expression in the presumptive branchial arch region coextensive with the EphA4-expressing neural crest. (b) A stage 23 embryo show in lateral view. EphA4-expressing neural crest has started to migrate, and the ventral domain has narrowed to a stripe at the rostral margin of the third arch. (c) A stage 26 embryo shown in lateral view. EphA4-expressing neural crest has migrated into the third branchial arch. (d) Coronal section through the ventral domain of EphA4 expression in the presumptive branchial arch region of a stage 19 embryo. Expression occurs in visceral mesoderm, and at lower levels in underlying endoderm. (e) Coronal section through ventral EphA4 expression domain in a stage 23 embryo. Expression is at low levels in visceral mesoderm, and at higher levels in a broader domain in the underlying endoderm. (f) Krox-20 gene expression in a stage 26 embryo. Expression in neural crest is restricted to the third branchial arch. (g–i) Expression pattern of EphB1. (g) At stage 25, EphB1 expression occurs in migrating third and fourth arch neural crest, and thus appears as a broader domain than EphA4. Expression also occurs in a ventral domain coextensive with the neural crest; the border of the neural crest is difficult to see because expression occurs at similar levels in the ventral region. (h) At stage 27, expression occurs in the fourth and more caudal branchial arches. (i) Section through stage 25 embryo. Expression is detected in neural crest, visceral mesoderm and endoderm. (j) Lateral view of EphA4 expression at stage 19 (see (a)), showing the ventral EphA4 expression domain in more detail. (k–m) Detection of EphA4 (strong staining in neural crest) plus EphB1 (weaker staining) expression during neural crest migration at stages 23 (k), 24 (l) and 26 (m). EphB1 expression extends caudal to EphA4 expression both in third arch neural crest and in the ventral domain. Abbreviations: r3/5, rhombomeres 3/5; n3/4, neural crest destined for third/fourth arch; v, ventral expression domain; fb, forebrain; cg, cement gland; pn, pronephros; ec, ectoderm; m, mesoderm; en, endoderm.
Figure 3. Effects of expressing truncated EphA4 and EphB1 on third arch neural crest. RNA encoding truncated EphB1 (a), EphA4, (b,e) or a mixture of these RNAs (c,f,h–l) was microinjected into one cell of two-cell stage Xenopus embryos. The embryos were allowed to develop to stage 21 (a–c), or stages 23–25 (d–l), then fixed and in situ hybridisation carried out. (a–c) Dorsal views of flat mounted embryos, with the injected side to the right in (a,b), and on the left in (c). (d,g) Uninjected side of injected embryos. (e,f,h–l) Injected side of embryos. Krox-20 gene expression was detected as a marker of third arch neural crest (a–f), Krox-20 plus EphA2 (Eck) to detect third arch (stronger signal) plus second arch (weaker signal) neural crest (g–i), or EphA2 alone (j). Krox-20-expressing neural crest cells were observed in ectopic locations; the extent of this disruption varied between embryos within an experiment and between different batches of embryos. To further analyse whether second arch neural crest was affected in embryos in which third arch crest migration was disrupted, DIG-labelled EphA2 probe (k) then fluorescein-labelled Krox-20 probe (l) were detected sequentially. The second arch stream is normal, but ectopic caudal third arch crest cells are detected; the appearance of this embryo has changed due to loss of pigmentation during processing. The white arrows indicate a normal restricted stream of third arch (n3) and second arch (n2) neural crest, and the black arrows indicate cells present in abnormal locations.
Figure 4. Expression of Xenopus ephrin-B2. (a–c) Expression of ephrin-B2 in stage 21 (a,b) and stage 26 (c) Xenopus embryos. Transcripts are detected in the eye, r2, r4 and r6, in neural crest migrating to the second branchial arch, and in a ventral domain coextensive with this neural crest. A coronal section through this ventral domain (b) reveals expression in mesoderm. (d–g) Double detection of ephrin-B2 plus EphA4 expression. At stage 21 (d,e), the expression domains in second and third arch crest are contiguous with each other, and by stage 24 (f,g), the migrating cells have become separated as they enter the forming branchial arches. Abbreviations: e, eye; others are as in the Figure 2 legend.
Figure 5. Effects of overexpressing ephrin-B2 on neural crest migration. RNA encoding ephrin-B2 was injected into Xenopus embryos at the two-cell stage, which were later fixed and in situ hybridisation carried out to detect second and third arch neural crest markers. (a,b) Embryos fixed at stage 20 and analysed with a Krox-20 probe. (c,d) Embryos fixed at stage 24 and analysed with a Krox-20 probe. (e,f) Embryos fixed at stage 20 and analysed with Krox-20 plus EphA2 probes. (g,h) Embryos fixed at stage 24 and analysed with an EphA2 probe. Ectopic Krox-20-expressing cells (black arrows) are detected caudal and/or rostral to the third arch neural crest stream (n3), which is often abnormal in shape. In contrast, the second arch crest (n2) marked by EphA2 expression appears normal. In some embryos (e), there was a gap between the third and second arch neural crest which could reflect a narrower stream of third arch crest cells.
Figure 6. Effects of targeted expression in neural crest. RNA was injected into the A1 or A2 blastomeres of 32-cell embryos in order to target expression of ephrin-B2 or truncated receptors to ectodermal derivatives, including branchial neural crest. Krox-20 expression was then detected by in situ hybridisation (blue signal). In some experiments, fluorescein-dextran was co-injected as a lineage tracer and detected using alkaline-phosphatase-conjugated, anti-fluorescein antibody and Fast Red substrate (red signal in (e–h)). (a–f) Embryos injected with RNA encoding truncated EphB1 plus EphA4 shown in lateral views (a–e) and in a coronal section (f) through the embryo shown in (e). The latter embryo has some non-specific staining that is often seen using Fast Red substrate. (g,h) Embryo injected with RNA encoding ephrin-B2 shown in a lateral view (g) and coronal section (h). Ectopic Krox-20-expressing cells are indicated by black arrows, and the lineage tracer in branchial neural crest by white arrows.