The structures of the face in vertebrates are largely derived from neural crest. There is some evidence to suggest that the form of the facial pattern is determined by the crest, and that it is specified before migration as to the structures that it is able to form. The neural crest is able to control the form of surrounding, non-neural crest tissues by an instructive interaction. Some of this cranial crest is derived from a region of the hindbrain that expresses Hox 2 homeobox genes in an overlapping and segment-restricted pattern.

We have found that neurogenic and mesenchymal neural crest expresses Hox 2 genes from its point of origin beside the neural plate, during migration and after migration has ceased and that rhombomeres 3 and 5 do not have any expressing neural crest beside them. Each branchial arch expresses a different combination or code of Hox genes in a segment-restricted way.

The surface ectoderm over the arches initially does not express Hox genes, and later adopts an expression pattern that reflects that of neural crest that has come to underlie it. We suggest that initially the neural plate and neural crest are spatially specified, while the surface ectoderm is unpatterned. Subsequently some positional information could be transferred to the surface ectoderm as a result of an interaction with the neural crest.

Given that the role of the homologous genes in insects is position specification, and that neural crest is imprinted before migration, we suggest that Hox 2 genes are providing part of this positional information to the neural crest and hence are involved in patterning the structures of the branchial arches.

The development of the mesoderm of the vertebrate head involves different mechanisms from that of the trunk. These differences may be related to the presence of a tissue, the neural crest, which is able to undertake roles that in the rest of the body are played by the paraxial mesoderm. Neural crest cells originate from the boundary between the neural plate and the surface ectoderm (Verwoerd and van Oostrom, 1979; Nichols, 1981). The crest migrates ventrally and in the hindbrain region of all vertebrates populates a series of repeating structures, the branchial arches (Morriss-Kay and Tan, 1987; Le Douarin, 1983; Noden, 1988; A. Lumsden, personal communication). In the trunk, there are subpopulations of crest that follow different temporal and spatial migration pathways, and it is possible that, in the head, there are different populations as well (Lumsden, 1989. The early migrating cells in the head reach the branchial arches and adopt a mesenchymal fate, while later emerging cells remain closer to the neural tube and adopt a neural fate (Nichols, 1981, 1986).

Interactions with head epithelia such as the neural tube and surface ectoderm seem to be important in neural crest differentiation and some aspects of head morphogenesis in chick (Bee and Thorogood, 1980; Thorogood et al. 1986; reviewed in Hall, 1987), Xenopus (Seufert and Hall, 1990) and mouse (Lumsden, 1987). However, evidence suggests that while the neural crest is extensively dependent on surrounding tissue to allow differentiation, some patterning information resides within the crest itself. This has been shown in experiments where sections of chick midbrain neural plate, whose crest normally colonises the first (maxillary and mandibular) arch, have been grafted to the second (hyoid) arch level (Noden, 1983, 1988). The neural crest migrated into the hyoid arch, but there formed mandibles, which are first arch structures. In addition, these ectopic mandibles had a set of muscles attached to them that were derived from the second arch paraxial mesoderm, but resembled first arch muscles (Noden, 1988). Duplicate beaks were also formed on the surface, suggesting that the differentiation pattern of second arch paraxial mesoderm and surface ectoderm was controlled by the neural crest.

Another aspect of head development that is unique is illustrated by hindbrain formation. Early in development the hindbrain is composed of a repeating pattern of bulges, the rhombomeres. Single cell marking in chick has shown that the rhombomeres are compartments that show lineage restrictions (Fraser et al. 1990), and conservation of their number and position in all vertebrates suggests that they are true segments which play an important part in the development of the head (for refs see Lumsden, 1990).

Much recent work has centred on the isolation of families of transcription factors potentially involved in early developmental decisions such as establishing the basic body plan. Important examples are the Drosophila (HOM-C) and vertebrate (Hox) Antennapedia class homeotic genes, which are organised in tight clusters (Akam, 1989). The extensive similarities between the clusters of these two organisms suggest that HOM-C and the Hox complexes are descendants of an ancestral cluster of genes, whose basic role, specification of position along the body axis, has been retained from a common ancestor (Graham et al. 1989; Duboule and Dollé, 1989; Beeman, 1987; Beeman et al. 1989; Akam, 1987; Kessel et al. 1990).

In Drosophila, boundaries of gene expression correlate with specific segments (Akam, 1987), which is also true of the rhombomeric segments of the vertebrate head. In the mouse hindbrain, cutoffs of Hox 2 genes correspond to rhombomeric boundaries, with successive genes showing expression limits separated by two rhombomere units (Wilkinson et al. 1989b; Murphy et al. 1989).

The association between branchial arches and specific rhombomeres are conserved in all vertebrates at similar stages of development. Because of the major contribution of neural crest to the branchial arches, and its ability to affect the development of surrounding tissues, the pattern of neural crest differentiation controls the structure of most of this region of the body. Given that the hindbrain crest is a candidate for a vertebrate tissue specified on the basis of its position along the rostrocaudal axis, and evidence of Hox 2 expression in the hindbrain and in neurogenic crest (Holland and Hogan, 1988; Graham et al. 1989; Wilkinson et al. 1989b; Frohman et al. 1990), we have examined the expression of Hox 2 genes in the neurogenic and mesenchymal neural crest of mouse embryos. We find that the limits of expression correspond to the branchial arches, and that each arch has a unique code of gene expression, related to the rhombomeric origin of the crest of that arch. These results suggest that Hox genes may have a role in patterning head structures beyond the neural plate.

In situ hybridisation and serial reconstruction was performed as described by Wilkinson and Green, 1990, except that probes were resuspended to a final activity of 1 × 104 cts min−1 μl−1 in hybridisation buffer The probes were identical to those of Wilkinson et al. 1989b for Hox 2.7 and

Hox 2.9, and Rubock et al (1990) for Hox 2.8 They were as follows

The mouse strains used were CBA and CBA/C57 crosses, and stage was determined on the basis of timed matmgs and otic vesicle morphology.

Serial sections were prepared from camera-luada drawings and reconstructed using the NIMR ‘ssrcon’ package.

Hox 2.8 expression in the branchial arches

In mouse, the neural crest begins to leave the margins of the neural plate at 8 days of development (4 somites) in the cranial region, and by days (11 somites), migration is well under way (Verwoerd and van Oostrom, 1979; Nichols, 1981). Initially we investigated the expression pattern in 9 day embryos in which the migration of cranial crest is complete, to see if there was a correlation between expression of Hox 2 genes in the hindbrain and neural crest.

Fig. 1 shows a series of coronal sections of a 9 day embryo hybridised with Hox 2.8. The planes of section through the embryo are shown in Fig. IE and the position of the otocyst adjacent to rhombomeres 5 and 6 allows orientation within the hindbrain. Fig. 1A shows hybridisation in the neural tube ending in rhombomere 3, and in the VII/VIII cranial ganglion complex, as previously described (Wilkinson et al. 1989b). Fig. IB is more ventral, and it is possible to see the beginnings of the first and second branchial arches, the latter adjacent to the cutoff of expression in the neural tube. Hybridisation is seen in the centre of arch 2 and in a more posterior domain, as well as within the neural tube itself. In Fig. 1C, arch 2 is visible as a discrete structure, and it is clear that there is no expression in arch 1, while arch 2 and more posterior regions express Hox 2.8. Fig. 1D crosses the heart, and is ventral to all of the arches except arch 1. Arch 1 is clearly not expressing Hox 2.8, while hybridisation is detected in posterior regions.

Fig. 1.

Expression of Hox 2.8 in coronal hindbrain sections of a 9 day mouse embryo. The relative position of sections is shown in E. A is the most dorsal section. The positions of the three branchial arches that have formed at this stage are shown in A and by the vertical bars on all sections, which indicate the interfaces between branchial arches. Note expression in section D in surface ectoderm terminates before the first arch; no first arch ectoderm expresses Hox 2 8. Sections C’and D’ are high power views of sections C and D respectively. (A–D ×80, C’ and D’ ×400).

Fig. 1.

Expression of Hox 2.8 in coronal hindbrain sections of a 9 day mouse embryo. The relative position of sections is shown in E. A is the most dorsal section. The positions of the three branchial arches that have formed at this stage are shown in A and by the vertical bars on all sections, which indicate the interfaces between branchial arches. Note expression in section D in surface ectoderm terminates before the first arch; no first arch ectoderm expresses Hox 2 8. Sections C’and D’ are high power views of sections C and D respectively. (A–D ×80, C’ and D’ ×400).

Hox 2.8 expression is confined to arch 2 and posterior in areas colonised by neural crest. The posterior expression is in a region that contains paraxial mesoderm, and it is not possible, because of the lack of cellular resolution with in situ hybridisation, to distinguish between tissues expressing Hox 2.8. The branchial arches are largely derived from neural crest, although there is also contribution from paraxial mesoderm in the core of the branchial arch in chick (Noden, 1988). This paraxial contribution to ventral parts of the arch is small and is confined to the core of the arch, so we suggest that the hybridisation seen in arch 2 is largely due to expression in the neural crest.

To confirm that Hox 2.8 expression was consistent with all areas colonised by cranial neural crest, a number of sections of a day embryo frontal to the branchial arches were hybridised and the embryo reconstructed. The reconstructed embryo is shown in Fig. 2D, with three of the sections used shown in Fig. 2A–C. The first and second branchial arches are clear as discrete structures in this plane of section, and the difference between the expression in first and second arches can be seen. Fig. 2A is the most dorsal section, and the structures in it are shown in Fig. 2A’. A cranial ganglion hybridises, and its position posterior of the otocyst but close to it suggests that it is the superior ganglion of the IX/X ganglion complex. Fig. 1A and previous work (Wilkinson et al. 1989b) suggest that Hox 2.8 is also expressed in the VII/VII ganglion complex, consistent with the idea that genes are expressed in all cranial ganglia posterior to the cutoff in the neural tube. There may be a slightly increased labelling over the otocyst, but we do not believe it to be labelled above background as it is a tissue more dense than the mesenchyme surrounding it. Fig. 2B and C show first and second branchial arches very clearly, and in both sections surface ectoderm of the second arch is positive in contrast to that of the first arch. In the reconstructed embryo, expression can be seen to be within all areas lateral of the neural tube colonised by the neural crest, and extending ventrally into the second branchial arch. There is no expression in the first arch at any level within the embryo, demonstrating that Hox 2.8 expression distinguishes the second from the first arch.

Fig. 2.

Serial reconstruction of Hox 2.8 expression in the head of a 912 day mouse embryo. The age of the embryo was established by the morphology of the otic vesicle. A – C show three sections used in the reconstruction, with A the most dorsal and C the most ventral. D is a reconstruction of the expression. The central nervous system is indicated in green. Expression in areas lateral of the hindbrain colonised in neural crest are shown filled in red. The otic vesicle is shaded in blue. Anterior is to the left. A’ is a drawing to illustrate structures in A. Hybridisation in neural crest is shown in red. Fb, forebrain; Bl, first branchial arch; NC, neural crest; Hb, hindbrain; OV, otic vesicle; glX, part of IX\X ganglion complex. (×100).

Fig. 2.

Serial reconstruction of Hox 2.8 expression in the head of a 912 day mouse embryo. The age of the embryo was established by the morphology of the otic vesicle. A – C show three sections used in the reconstruction, with A the most dorsal and C the most ventral. D is a reconstruction of the expression. The central nervous system is indicated in green. Expression in areas lateral of the hindbrain colonised in neural crest are shown filled in red. The otic vesicle is shaded in blue. Anterior is to the left. A’ is a drawing to illustrate structures in A. Hybridisation in neural crest is shown in red. Fb, forebrain; Bl, first branchial arch; NC, neural crest; Hb, hindbrain; OV, otic vesicle; glX, part of IX\X ganglion complex. (×100).

Hox 2.8 is expressed in specific regions of surface ectoderm

The areas of surface ectoderm lateral to the edges of the neural plate are known to produce thickenings or placodes, which generate neural derivatives (D’Amico-Martel and Noden, 1983; Le Douarin et al. 1986). In the light of this and the recent work of Couly and Le Douarin (1990) on the contributions of ectoderm lateral of the neural plate to the head, we were interested to see the extent of Hox 2 expression in the surface ectoderm. In Fig. 1C’ (a high power view of Fig. 1C) and Fig. 2B and 2C, expression of Hox 2.8 is seen on the surface of the second branchial arch, and not on the surface of the first arch. This is confirmed in Fig. 1D’, where surface ectoderm of arch 1 has no signal, but the adjacent posterior ectodermal tissue expresses very strongly. This demonstrates that the surface ectoderm expresses Hox 2.8 at 9 days in the same way as the underlying crest mesenchyme.

Expression of Hox 2.8 in crest at time of emergence

To determine the timing of the onset of Hox 2.8 expression in the crest-derived mesenchyme, we investigated the expression in neural groove stage (8i day) mouse embryos in transverse section. A series of sections are shown in Fig. 3, where the plane is not quite perpendicular to the long axis of the embryo; thus one side of the neural plate is more anterior than the other. Sections in the series more anterior than those shown here showed no labelling above background. In the most anterior section (Fig. 3A), only the neural plate is labelled, and only on one side. This section is at the level of the anterior cutoff of expression, between rhombomere 2 (left side, not expressing) and rhombomere 3 (right side, expressing). A near adjacent, more posterior section (Fig. 3B) shows labelling of both sides of the neural plate, with no expression in the other embryonic parts of the section. Fig. 3C shows an additional patch of expression lateral to the dorsal edge of the neural plate. This additional site of labelling is in neural crest (Fig. 3C; nc) (Nichols, 1981, 1986; Hall, 1987; Noden, 1987) and is continuous with the neural plate extending ventrolaterally first on one side and then both sides (data not shown, Fig. 3C). Fig. 3D shows a second region on the right of the section where only neural plate is labelled. We interpret the sections shown to be through the rhombomeres indicated in the diagram on the right of the figure. It indicates both rhombomeres that produce neural crest, and areas that correspond to crest-free zones, based on Hox 2.8 expression and previous studies (Anderson and Meier, 1981; Tan and Morriss-Kay, 1985; Wilkinson et al. 1989b). Thus it seems that where neural crest does arise it expresses Hox 2 genes from time of emergence and that the neural crest migrating into the arches has a Hox 2 label or code.

Fig. 3.

Expression of Hox 2 8 in serial, transverse sections of an 812 day mouse embryo hindbrain. The dorsal surface that will become the ventricular surface of the neural tube is uppermost in all sections. A is the most anterior section. The sections are slightly oblique, such that the left-hand side of each section is more anterior than the right-hand side The arrowheads indicate extraembryomc membranes that are expressing Hox 2.8, and serve as a positive control for hybridisation. At this stage of mouse development, the rhombomeres are only beginning to be visible as morphological units, although their future boundaries are clearly indicated by patterns of gene expression. The areas where neural plate alone expresses Hox 2.8 would correspond to areas suggested to be crest free from other work (Anderson and Meier, 1981; Tan and Morriss-Kay, 1985; Lumsden and Sprawson, 1991). In the light of this, and the known position of the cutoff of Hox 2.8 expression at the r2/r3 boundary (Wilkinson et al. 1989b), we interpret the sections shown to be through the rhombomeres indicated on the diagram on the right of the figure, which also indicates which rhombomeres are producing neural crest. The tissues known to be expressing Hox 2.8 are shaded m grey stipple, and the lobes lateral to the neural plate indicate areas of neural crest that are known to be produced by particular lengths of neural plate, np, neural plate; nc, neural crest; nf, neural fold; ect, surface ectoderm; eem, extraembryonic membrane; fp, floorplate (×200).

Fig. 3.

Expression of Hox 2 8 in serial, transverse sections of an 812 day mouse embryo hindbrain. The dorsal surface that will become the ventricular surface of the neural tube is uppermost in all sections. A is the most anterior section. The sections are slightly oblique, such that the left-hand side of each section is more anterior than the right-hand side The arrowheads indicate extraembryomc membranes that are expressing Hox 2.8, and serve as a positive control for hybridisation. At this stage of mouse development, the rhombomeres are only beginning to be visible as morphological units, although their future boundaries are clearly indicated by patterns of gene expression. The areas where neural plate alone expresses Hox 2.8 would correspond to areas suggested to be crest free from other work (Anderson and Meier, 1981; Tan and Morriss-Kay, 1985; Lumsden and Sprawson, 1991). In the light of this, and the known position of the cutoff of Hox 2.8 expression at the r2/r3 boundary (Wilkinson et al. 1989b), we interpret the sections shown to be through the rhombomeres indicated on the diagram on the right of the figure, which also indicates which rhombomeres are producing neural crest. The tissues known to be expressing Hox 2.8 are shaded m grey stipple, and the lobes lateral to the neural plate indicate areas of neural crest that are known to be produced by particular lengths of neural plate, np, neural plate; nc, neural crest; nf, neural fold; ect, surface ectoderm; eem, extraembryonic membrane; fp, floorplate (×200).

In this series of sections at days no’ labelling above background was detected in surface ectoderm, although it is continuous with the neural plate that expresses high levels of Hox 2.8. Adjacent extraembryonic membranes showed intense labelling, indicating that tissue thickness per se was not the cause of lack of signal (Fig. 3C). Therefore the high level of expression seen in surface ectoderm at 9 days appears to represent a new site of expression. These sections also show that there is no detectable expression in non-crest mesenchyme, consistent with the idea that neural crest and plate, not mesoderm, are the major sites of expression at this stage.

Other genes of Hox 2 are expressed in specific branchial arches

Fig. 4 shows the expression pattern of Hox 2.7 and Hox 2.9 in the branchial arches at similar levels of the body of a day mouse embryo to those shown for Hox 2.8. The embryo shown in Fig. 4A and C has been hybridised with Hox 2.7, while that shown in Fig. 4B and D has been hybridised with Hox 2.9.

Fig. 4.

Expression of other members of Hox 2 in the neural crest. A and C show sections of an embryo hybridised with Hox 2.7; B and D show sections of a different embryo hybridised with Hox 2 9 Anterior is in all cases to the left. A and B are more dorsal sections, running through the neural tube. The plane of section C is such that the top side is more dorsal than the bottom side of the section; the lower side runs through the branchial arches. Section D is more oblique, the top side is at the level of the otocyst, while the bottom is more ventral, running through the branchial arches. r4,r5,r6, rhombomere 4 etc.; g vii/viii, ganglia complex of the facial/acoustic nerve; b1 – b4, lst –4th branchial arch; nt, neural tube; ov, otic vesicle. (×100)

Fig. 4.

Expression of other members of Hox 2 in the neural crest. A and C show sections of an embryo hybridised with Hox 2.7; B and D show sections of a different embryo hybridised with Hox 2 9 Anterior is in all cases to the left. A and B are more dorsal sections, running through the neural tube. The plane of section C is such that the top side is more dorsal than the bottom side of the section; the lower side runs through the branchial arches. Section D is more oblique, the top side is at the level of the otocyst, while the bottom is more ventral, running through the branchial arches. r4,r5,r6, rhombomere 4 etc.; g vii/viii, ganglia complex of the facial/acoustic nerve; b1 – b4, lst –4th branchial arch; nt, neural tube; ov, otic vesicle. (×100)

In Fig. 4A and C, the plane of section is slightly oblique, the lower side of the embryo being more ventral than the upper. Fig. 4A shows the cutoff of expression in the neural tube at the r4/r5 boundary, as shown by Wilkinson et al. 1989b. Fig. 4C shows a more ventral section of the same embryo. The second and third arches are visible only on the upper side of the embryo because of the tilted plane of section. Hybridisation is clearly excluded from the first two branchial arches, and confined to the third and posterior. Hox 2.6 shows an identical background level of expression over the first three branchial arches (data not shown), with a more intense region of expression lateral of the neural tube and posterior of the branchial arches.

Fig. 4B shows a dorsal level of the neural tube. There are three very clear areas of Hox 2.9 hybridisation; the neural tube in rhombomere 4, and either side in the facial-acoustic cranial ganglion complex. This complex is composed of three lobes, which are particularly clear on the upper side of the figure. Three lobes can also be seen in Fig. 1A hybridised with Hox 2.8. Thus both Hox 2.9 and Hox 2.8 label all parts of this ganglion complex. The neurons here are largely derived from the otic placode, but the support cells are a neural crest derivative.

Fig. 4D is a more ventral section of the same embryo, hybridised with Hox 2.9. It is tilted, so that on the upper side the otic vesicle and part of the labelled ganglion complex can be seen, while the lower side is deeper in the embryo, running through the level of the branchial arches. It is clear that while there is specific hybridisation on other parts of this section, there is no expression above background levels in either the first or second branchial arches. This is despite the fact that the second arch is populated by neural crest derived from rhombomere 4, which expresses Hox 2.9. This difference between mesenchymal and neurogenic crest agrees with the observations of Frohman et al. 1990.

The cranial neural crest seems to be specified before migration as to the structures it will form, and controls the structures formed by other, non-neural crest tissues (Noden, 1983, 1986). Hox genes in developing systems are thought to be one component of the process of assigning different states to otherwise equivalent groups of cells. The maintenance of a state may be manifested by the continued expression of these genes. We have shown that the neural crest arising from the hindbrain region carries a Hox 2 positional label or combinatorial code from its point of origin to end point of migration. Each branchial arch has a distinct code of Hox 2 expression (with arch one not expressing any Hox gene), and this arch-specific Hox 2 pattern is in the neural crest before it has reached the branchial arches. The cranial ganglia lateral of the neural tube also show the same pattern of expression as the neural tube itself. The expression patterns that we have found are summarised in Fig. 5. Given that Antennapedia class homeobox genes act as positional specifiers (Akam, 1987; Beeman, 1987; Beeman et al. 1989; Kessel et al.1990), a specific combination of Hox 2 expression could provide part of the molecular mechanism for imprinting of cranial neural crest.

Fig. 5.

Summary of Hox 2 expression found in the hindbrain and branchial arches. The stipple indicates the areas of neural plate where neural crest is produced, and the branchial arch into which it migrates. The ganglion next to rhombomere 2 is the V or trigermnal, that next to rhombomere 4 is the VII/VI or acoustic-facial complex, and those next to rhombomere 6 are the combined superior ganglia of the IX and X cranial nerves. The shading patterns shown in the cranial ganglia indicate expression of a combination of genes, and do not imply that there is spatial restriction of gene expression within a ganglion. The position in the Hox 2 cluster of the genes discussed in the text are shown at the bottom of the figure.

Fig. 5.

Summary of Hox 2 expression found in the hindbrain and branchial arches. The stipple indicates the areas of neural plate where neural crest is produced, and the branchial arch into which it migrates. The ganglion next to rhombomere 2 is the V or trigermnal, that next to rhombomere 4 is the VII/VI or acoustic-facial complex, and those next to rhombomere 6 are the combined superior ganglia of the IX and X cranial nerves. The shading patterns shown in the cranial ganglia indicate expression of a combination of genes, and do not imply that there is spatial restriction of gene expression within a ganglion. The position in the Hox 2 cluster of the genes discussed in the text are shown at the bottom of the figure.

The relationship between expression in neural tube and branchial arches

The pattern of expression in the rhombomeres seems to be out of phase with that of the branchial arches; Hox 2.8 is expressed in rhombomere 3, some of whose motor neurons contribute to cranial nerve V (the trigeminal) innervating arch 1, while in the arch mesenchyme it is confined to arch 2 and posterior. Our data at 8i days show that rhombomere 3 and probably r5 do not have hybridising material lateral of them, while the neural plate expresses strongly. This raises the possibility that some areas of the neural plate do not produce neural crest. Consistent with this, Krox 20, which is expressed in neural crest-derived boundary cap cells along the entire neuraxis, is not expressed lateral to rhombomeres 3 and 5 (Wilkinson et al. 1989a). SEM studies of chick and rat embryos at the time of crest emigration suggest that areas of neural tube are crest free (Anderson and Meier, 1981; Tan and Morriss-Kay, 1985). A definitive proof comes from dye injections at the dorsal midline of chick neural tubes (A. Lumsden, personal communication), which confirm that rhombomeres 3 and 5 do not produce any neural crest and that rhombomere 4 contributes neural crest to the whole of the second arch and no other. Together these data imply that the branchial arch expression that we see is out of phase with the neural tube because of the presence of crest-free rhombomeres. The lack of extensive mixing between different populations of neural crest along the rostrocaudal axis (A. Lumsden, personal communication) would mean that during migration into an arch relative spatial positions of cells are maintained, and hence pattern of gene expression.

Therefore the neural crest in an arch expresses a Hox 2 code related to its level of origin along the margins of the neural plate as shown in Fig. 5.

Patterning of facial ectoderm

Recently Couly and Le Douarin (1990) have investigated the fate of cells in this region of the body. At an early stage the surface ectoderm, prospective neural crest and neural plate are continuous, and at this time they made an orthotopic graft of marked (quail) cells, to identify the location of their descendants. On the basis of this they suggest that the neural tube, neural crest and surface ectoderm that will cooperate to form an arch all arise from the same axial level, and that all three have been initially specified as an ‘ectomere’ on the basis of their axial position.

As long as each tissue has some mechanism for specifying its axial position, the mechanism that specifies the position of part of the neuroepithelium, and hence the branchial arch it belongs to, need not be the same as that which specifies the position of the structures of the arch that it innervates. Thus the nerves and other structures of the same branchial arch need not have the same pattern of Hox 2 expression.

The evidence presented here suggests that a branchial arch is made up of a combination of structures with different positional values in terms of Hox gene expression. For example, rhombomere 3, the most anterior expressing Hox 2.8, gives rise to neurons that contribute to the trigeminal nerve, which innervates the first branchial arch (Lumsden and Keynes, 1989), but neither first arch mesenchyme nor surface ectoderm expresses Hox 2.8.

At days the surface ectoderm does not express Hox 2.8 above background levels, while at 9 days, when neural crest migration at this level is complete, the surface expresses Hox 2.8 in a similar way to the underlying mesenchyme. Given the evidence suggesting that mesenchyme can change the fate of overlying ectoderm (Noden, 1988; Richman and Tickle, 1989), perhaps the ectoderm pattern is a result of an interaction with underlying crest mesenchyme, rather than expression maintained from early stages of development. We suggest that initially the neural plate and the neural crest are spatially specified, while the branchial arch identity of surface ectoderm is unspecified. Subsequently some positional information could be transferred to the surface ectoderm as a result of an interaction with the neural crest.

The mechanism of head segmentation

The neural crest migrates from the neural plate, and so it is conceivable that, by patterning the neuroepithelium, Hox 2 genes are part of the process specifying the structures of the head and neck. Given the two segment periodicity of Hox 2 expression, it will be of interest to see how genes of the other three mammalian Hox clusters are expressed here, and whether there is a similar correlation between rhombomere expression and branchial arch coding.

The genes that we have described are unlikely to provide information such as A-P polarity within an arch, as they are homogenously expressed there. Information in the head region for skeletal morphogenesis must also come from the crest environment. This is supported by grafts of neural plate ip normal and reversed rostrocaudal orientation, in which the structures that form in the second arch are of normal rostrocaudal orientation (Noden, 1983). It is important to note that not all properties of cranial crest are consistent with regional identity being imprinted before migration. McKee and Ferguson (1984) extirpated mesencephalic crest, but found no resulting facial abnormalities, as crest anterior and posterior of the lesion migrated in to fill the defect. This may reflect differences in properties between branchial and more anterior crest, or that crest may become respecified.

The number and size of the repeating units in neural tube and branchial arches is probably established before Hox 2 expression reaches these regions. The neural crest does not appear to be intrinsically segmented despite arising from a segmented structure. Experiments in amphibia involving removal of pharyngeal endoderm, which reduces the number of branchial arches, have shown that the neural crest then migrates down to fill the reduced number of arches that are available (Balinsky, 1981), suggesting that the environment is causing the neural crest to form a series of repeated structures, rather than any intrinsic property of the crest such as its pattern of gene expression. The hindbrain and branchial arches develop in a very similar way in all vertebrates (Romer, 1971; Hall and Horstadius, 1988), and are thought to have been of great importance to primitive vertebrates (Gans, 1989; Langille and Hall, 1989). As the evolution of the neural crest has been suggested to be the key step separating the vertebrates from the other chordates (Gans and Northcutt, 1983), the expression pattern in the anterior regions of non-vertebrate chordates (Holland, 1990), which possess a nervous system but no neural crest, may give insights into how changes in gene expression correlate with the evolution of new aspects of the body plan.

We would like to thank Andrew Lumsden and Malcolm Maden for advice and suggestions throughout this work, and for communicating data before publication. We thank Leanne Wiedemann, Mai-Har Sham, Martyn Cook, Stefan Nonchev, Romita Das Gupta, Patrizia Ferretti and Peter Thorogood for comments and suggestions on the manuscript. This work was funded by the MRC and P. N. Hunt is in receipt of an MRC studentship.

Akam
,
M
(
1987
)
The molecular basis for metameric pattern in the Drosophila embryo
Development
101
,
1
22
.
Akam
,
M.
(
1989
)
Hox and HOM homologous gene clusters in insects and vertebrates
Cell
57
,
347
349
.
Anderson
,
C
and
Meier
,
S.
(
1981
)
The influence of the metameric pattern in the mesoderm on migration of cranial neural crest cells in the chick embryo
Devi Biol
85
,
385
402
Balinsky
,
B I
(
1981
).
An Introduction to Embryology, Fifth edn New York Saunders College Publishing
, (pp
465
—466)
Bee
,
J
and
Thorogood
,
P
(
1980
)
The role of tissue interactions in the skeletogemc differentiation of avian neural crest cells
Devi Biol.
78
,
47
62
Beeman
,
R.
(
1987
).
A homeotic gene cluster in the red flour beetle
.
Nature
327
,
247
249
Beeman
,
R.
,
Stuart
,
J.
,
Haas
,
M
and
Denell
,
R
(
1989
).
Genetic analysis of the homeotic gene complex (HOM-C) in the beetle
Tribolium castaneum Devi Biol
133
,
196
209
Couly
,
G
and
Le Douarin
,
N.
(
1990
)
Head morphogenesis m embryonic avian chimeras evidence for a segmental pattern in the ectoderm corresponding to the neuromeres
Development
108
,
543
558
D’amico-martel
,
A
and
Noden
,
D
(
1983
)
Contributions of placodal and neural crest cells to avian cranial peripheral ganglia
Am J. Anat
166
,
445
468
Duboule
,
D
and
Dollé
,
P
(
1989
)
The structural and functional organization of the munne HOX gene family resembles that of
Drosophila homeotic genes EMBO
8
,
1497
1505
Fraser
,
S
,
Keynes
,
R
and
Lumsden
,
A
(
1990
)
Segmentation m the chick embryo hindbrain is defined by cell lineage restrictions
Nature
344
,
431
435
Frohman
,
M
,
Boyle
,
M
and
Martin
,
G
(
1990
).
Isolation of the mouse Hox-2 9 gene, analysis of embryonic expression suggests that positional information along the anterior-posterior axis is specified by mesoderm
Development
110
,
589
607
Gans
,
C
(
1989
)
Stages in the origin of vertebrates analysis by means of scenarios
Biol Rev
64
,
221
268
Gans
,
C
and
Northcutt
,
R
(
1983
)
Neural crest and the origin of vertebrates, a new head
Science
220
,
268
274
Graham
,
A
,
Papalopulu
,
N
and
Krumlauf
,
R.
(
1989
).
The munne and Drosophila homeobox clusters have common features of organisation and expression
Cell
57
,
367
378
.
Graham
,
A
,
Papalopulu
,
N
,
Lorimer
,
J.
,
Mcvey
,
J
,
Tudenham
,
E.
and
Krumlauf
,
R
(
1988
)
Characterisation of a munne horneo box gene, Hox 2 6, related to the Drosophila Deformed gene
.
Genes Dev
2
,
1424
1438
Hall
,
B
(
1987
)
Tissue Interactions in Head Development and Evolution
In
Developmental and Evolutionary Aspects of the Neural Crest
(ed
P F A
Maderson
), pp
215
-
259
New York John Wiley
Hall
,
B.
and
Horstadius
,
S
(
1988
)
The Neural Crest
Oxford Oxford University Press
Holland
,
P
(
1990
)
Homeobox Genes and Segmentation Co-option, Co-evolution and Convergence
. In
Seminars in Developmental Biology The Evolution of Segmental Patterns,
vol.
1
,
issue 2
(ed. C. Stem), pp. 135-145 Philadelphia’ W B Saunders
Holland
,
P
and
Hogan
,
B.
(
1988
).
Expression of horneo box genes dunng mouse development’ a review
Genes Dev
2
,
773
7S2
.
Kessel
,
M.
,
Balling
,
R
and
Gruss
,
P
(
1990
)
Variations of cervical vertebrae after expression of a Hox 1 1 transgene in mice
Cell
61
,
301
308
Langille
,
R
and
Hall
,
B
(
1989
)
Developmental processes, developmental sequences and early vertebrate phylogeny
.
Biol Rev
64
,
73
91
Le Douarin
,
N.
(
1983
).
The Neural Crest
Cambridge
Cambridge University Press
Le Douarin
,
N.
,
Fontaine-Perus
,
J
and
Couly
,
G.
(
1986
).
Cephalic ectodermal placodes and neurogenesis
TINS
9
,
175
180
Lumsden
,
A
(
1987
)
The Neural Crest Contribution to Tooth Development in the Mammalian Embryo
. In
Developmental and Evolutionary Aspects of the Neural Crest
(ed
P F A
Maderson
), pp.
261
300
New York
.
John Wiley
Lumsden
,
A.
(
1989
)
Multipotent cells in the avian neural crest
.
TINS
12
,
81
83
Lumsden
,
A
(
1990
)
The Development and Significance of Hindbrain Segmentation
. In
Seminars in Developmental Biology The Evolution of Segmental Patterns,
vol
1
, issue
2
(ed.
C
Stem
), pp.
117
-
125 Philadelphia
. W B Saunders
Lumsden
,
A
and
Keynes
,
R
(
1989
).
Segmental patterns of neuronal development in the chick hindbrain
Nature
337
,
424
428
Mckee
,
G
and
Ferguson
,
M.
(
1984
)
The effects of mesencephalic neural crest cell extirpation on the development of chicken embryos
J. Anat
139
,
491
512
Morriss-Kay
,
G.
and
Tan
,
S S
(
1987
)
Mapping cranial neural crest migration pathways in mammalian embryos
.
TIG
3
,
257
261
.
Murphy
,
P
,
Davidson
,
D
and
Hill
,
R
(
1989
)
Segment-specific expression of a homeobox-containing gene in the mouse hindbrain
Nature
341
,
156
159
.
Nichols
,
D
(
1981
)
Neural crest formation in the head of the mouse embryo as observed using a new histological technique
JEEM
64
,
105
120
Nichols
,
D.
(
1986
)
Formation and distribution of neural crest mesenchyme to the first pharangeal arch region of the mouse embryo
Am. J Anat
176
,
221
231
.
Noden
,
D M
(
1983
)
The role of the neural crest in patterning of avian cranial skeletal, connective, and muscle tissues
Devi Biol.
96
,
144
165
Noden
,
D. M
(
1986
)
Patterning of avian craniofacial muscles
Devi Biol
116
,
347
356
.
Noden
,
D M.
(
1987
)
Interactions Between Cephalic Neural Crest and Mesodermal Populations
. In
Developmental and Evolutionary Aspects of the Neural Crest
(ed.
P F A
Maderson
), pp
89
-
119
.
New York John Wiley
Noden
,
D. M.
(
1988
)
Interactions and fates of avian craniofacial mesenchyme
.
Development 103 Supplement, 121-140
.
Richman
,
J
and
Tickle
,
C
(
1989
).
Epithelia are interchangeable between facial pnmordia of chick embryos and morphogenesis is controlled by the mesenchyme
Devi Biol
136
,
201
210
.
Romer
,
A
(
1971
)
The Vertebrate Body, Shorter Version, Fourth edn Philadelphia: W B. Saunders Company
, (pp
163
167
)
Rubock
,
M
,
Larin
,
Z
,
Cook
,
M
,
Papalopulu
,
N
,
Krumlauf
,
N
and
Lehrach
,
H
(
1990
)
A yeast artificial chromosome containing the mouse homeobox cluster Hox-2
Proc natn Acad Sci USA
87
,
4751
4755
Seufert
,
D
and
Hall
,
B
(
1990
).
Tissue interactions involving cranial neural crest in cartilage formation in Xenopus laevis (Daudin)
Cell Diff Devel
32
,
153
166
Tan
,
S S.
and
Morrjss-Kay
,
G.
(
1985
)
The development and distribution of the cranial neural crest in the rat embryo
Cell Tissue Res
240
,
403
416
Thorogood
,
P
,
Bee
,
J
and
Von Der Mark
,
K
(
1986
)
Transient expression of collagen type II at epithehomesenchymal interfaces during morphogenesis of the cartilaginous neurocranium
Devl Biol
116
,
497
509
Verwoerd
,
C
and
Van Oostrom
,
C
(
1979
)
Advances in Anatomy, Embryology and Cell Biology, 58 Cephalic Neural Crest and Placodes
Berlin Springer-Verlag
Wilkinson
,
D. G.
,
Bhatt
,
S
,
Chavrier
,
P
,
Bravo
,
R
and
Charnay
,
P
(
1989a
)
Segment-specific expression of a zinc finger gene in the developing nervous system of the mouse
.
Nature
337
,
461
464
Wilkinson
,
D G
,
Bhatt
,
S
,
Cook
,
M
,
Boncinelli
,
E
and
Krumlauf
,
R
(
1989b
)
Segmental expression of hox 2 homeobox-containing genes in the developing mouse hindbrain
Nature
341
,
405
409
Wilkinson
,
D G
and
Green
,
J
(
1990
).
In situ hybridization and three-dimensional reconstruction of serial sections
In
The Practical Approach Series Postunplantation Mouse Embryos A Practical Approach
(ed
D
Rickwood
and D.
L
Cockcroft
), pp
155
171
Oxford IRL Press