EVOLUTION & DEVELOPMENT

8:2, 130 –141 (2006)

Expression of Dlx genes during the development of the zebrafish pharyngeal dentition: evolutionary implications V. Borday-Birraux,a,1 C. Van der heyden,b,1 M. Debiais-Thibaud,a L. Verreijdt,b D. W. Stock,c A. Huysseune,b and J.-Y. Sired, a

PGE UPR 9034, CNRS, Gif-sur-Yvette, France Biology Department, Ghent University, Belgium c Department of Ecology and Evolutionary Biology, University of Colorado, Boulder, CO 80309 USA d Universite´ Paris 6, CNRS UMR 7138, Case 7077, 7, Quai St. Bernard, Paris cedex 05, France b

Author for correspondence (email: [email protected]) 1

Both authors contributed equally to this work.

SUMMARY In order to investigate similarities and differences in genetic control of development among teeth within and between species, we determined the expression pattern of all eight Dlx genes of the zebrafish during development of the pharyngeal dentition and compared these data with that reported for mouse molar tooth development. We found that (i) dlx1a and dlx6a are not expressed in teeth, in contrast to their murine orthologs, Dlx1 and Dlx6; (ii) the expression of the six other zebrafish Dlx genes overlaps in time and space, particularly during early morphogenesis; (iii) teeth in different locations and generations within the zebrafish dentition differ in the number of genes expressed; (iv) expression similarities and

differences between zebrafish Dlx genes do not clearly follow phylogenetic and linkage relationships; and (v) similarities and differences exist in the expression of zebrafish and mouse Dlx orthologs. Taken together, these results indicate that the Dlx gene family, despite having been involved in vertebrate tooth development for over 400 million years, has undergone extensive diversification of expression of individual genes both within and between dentitions. The latter type of difference may reflect the highly specialized dentition of the mouse relative to that of the zebrafish, and/or genome duplication in the zebrafish lineage facilitating a redistribution of Dlx gene function during odontogenesis.

INTRODUCTION

In the zebrafish, Danio rerio, a polyphyodont- and heterodont (sensu lato) species, molecular studies on tooth development are only emerging (Payne et al. 2001; Hsiao et al. 2002; Jackman et al. 2004; Laurenti et al. 2004). Teeth are found in the pharyngeal region only (Huysseune et al. 1998). Their development is well known at the tissue and cell level (Huysseune et al. 1998, and see review in Sire and Huysseune 2003), and the pattern of tooth formation (i.e., the position and order of tooth appearance) and replacement has been described in detail (Van der heyden and Huysseune 2000; Van der heyden et al. 2000). Although actinopterygian and mammalian lineages have evolved separately for approximately 420 million years, resemblances between zebrafish and tetrapod tooth development have prompted us to question whether zebrafish can be used as a model for vertebrate tooth development also at the genetic level. In addition, a comparison between zebrafish and mouse provides an estimate of what could be considered an upper limit of divergence in mechanisms of tooth development, as the comparison involves teeth from different germ

A fundamental question in evolutionary developmental biology is how development evolves to produce homologous organs of different form (Rudel and Sommer 2003). The vertebrate dentition provides a wealth of opportunities to address this question through the identification of developmental genetic similarities and differences between both iterative homologs (e.g., teeth in different positions or belonging to different generations within a species) and historical homologs (equivalent teeth between species). Evolutionary analyses of teeth at the molecular level are still rare (Kera¨nen et al. 1998; Fraser et al. 2004; Jackman et al. 2004) mostly because the data available so far come almost exclusively from the mouse (cf. Nieminen et al. 1998; Stock 2001; and the website http:// bite-it.helsinki.fi). Yet, the mouse dentition is highly specialized in that there is only one tooth generation (monophyodont), several tooth positions have been lost and some teeth (the incisors) are continuously growing (Huysseune and Sire 1998). 130

& 2006 Blackwell Publishing No claim to original US government works

Borday-Birraux et al.

Dlx expression in zebra¢sh teeth

layers (endoderm in the zebrafish vs. ectoderm in the mouse), developing in a different molecular environment (Hox-expressing in the zebrafish vs. non-Hox-expressing in the mouse), in different locations (restricted to the ventral parts of the visceral arch in zebrafish, in contrast to mouse), and in distant species that are moreover separated by a genome duplication. In the present study, we take advantage of the well-characterized genome of the zebrafish to address the extent to which all of the members of a particular gene family have diversified in expression among homologous teeth. The specific family examined, the Dlx genes, were chosen because (i) their expression pattern during tooth morphogenesis and differentiation is well known in humans and mice (Weiss et al. 1994; Thomas et al. 1997, 2000; Bei and Maas 1998; Price et al. 1998a, b; Zhao et al. 2000), (ii) mutations of Dlx genes have been reported to lead to severe disorders in the human dentition (e.g., DLX 3: Price et al. 1998a, b), and (iii) the composition of the Dlx gene family is well known in the main vertebrate lineages (Stock et al. 1996; Neidert et al. 2001;

Table 1. Nomenclature of the Dlx genes in mouse and zebrafish allowing ortholog identification (after Panganiban and Rubenstein 2002) (A). Linkage relationships of zebrafish Dlx and Hox genes (after Amores et al. 1998) (B)

Mouse Dlx1 Dlx2 Dlx3 Dlx4 (formerly Dlx7) Dlx5 Dlx6

A

Zebrafish: Former Names

dlx1a dlx2a dlx2b dlx3b dlx4a

Dlx1 Dlx2 Dlx5 Dlx3 Dlx8

dlx4b dlx5a dlx6a

Dlx7 Dlx4 Dlx6

hox aa

dlx6a

hox ba

dlx4a

hox bb

dlx4b

hox da B

Zebrafish: New Names

dlx1a

131

Stock 2005). The Dlx genes encode homeodomain-containing transcription factors that are closely related to those of the Drosophila Distal-less (Dll) gene and are found in all chordate phyla. Six Dlx genes are known in humans (Simeone et al. 1994; Scherer et al. 1995; Price et al. 1998a, b) and mice (Robinson and Mahon 1994; Weiss et al. 1994, 1995; Nakamura et al. 1996; Stock et al. 1996; Panganiban and Rubenstein 2002), whereas eight are found in the zebrafish genome (Ekker et al. 1992; Akimenko et al. 1994; Stock et al. 1996; Ellies et al. 1997; Panganiban and Rubenstein 2002) (Table 1A). In the mouse, Dlx genes are found in three pairs of convergently transcribed genes (Dlx1–2, 3–4 and 5–6). In the zebrafish, orthologs are similarly arranged (dlx1a–2a, 3b– 4b, 5a–6a), but two additional genes, dlx2b and dlx4a, are not linked to another Dlx gene and are located on two distinct chromosomes (Ellies et al. 1997; Amores et al. 1998) (Table 1B). In some cases, the two genes that constitute a pair have overlapping expression patterns because of the presence in the intergenic region of two similar regulatory regions (Qiu et al. 1997; Zerucha et al. 2000; Robledo et al. 2002; Solomon and Fritz 2002), and can compensate for the loss of one another. Moreover, species differences in ortholog function have been reported (e.g., Quint et al. 2000). In the present article, we analyze the expression pattern of all eight Dlx genes during the development of first-generation

dlx5a

dlx3b

dlx2a dlx2b

Fig. 1. Schematic representation of the tooth pattern during development of zebrafish from 48 to 168 hours postfertilization at 28.51C. During this period, teeth develop in three positions only (3V, 4V, and 5V), with two tooth generations (V1 and V2). The hours are only indicative of a mean developmental stage, which could vary depending on various environmental and genetic factors.

132

EVOLUTION & DEVELOPMENT

Vol. 8, No. 2, March^April 2006 48 hpf with position 4V (tooth 4V1, superscript indicating the generation number of the tooth), followed by teeth 3V1 and 5V1 at 56 hpf. At 80 hpf, tooth 4V1 is fully developed and attached to the bone support. Teeth 3V1 and 5V1 are attached at 144 hpf. The first replacement tooth appears in position 4V (4V2) at 80 hpf, followed by replacement teeth 3V2 and 5V2 at 144 hpf. Teeth 4V2, 3V2, and 5V2 are fully formed and attached at 12 days postfertilization (dpf) (not shown). The other positions appear later during ontogeny, from 12 dpf onwards.

Odontogenetic stages

Fig. 2. Schematic comparison of tooth development phases in a zebrafish pharyngeal tooth (A) and in a mouse molar (B). EM, early morphogenesis ( 5 initiation1bud in mice); LM, late morphogenesis ( 5 cap); ED, early cytodifferentiation ( 5 early bell); LD, late cytodifferentiation ( 5 late bell). The final phase, attachment, is not shown because Dlx expression was not reported in mice after late cytodifferentiation. (A) modified after Huysseune et al. (1998); (B) modified after Zhao et al. (2000).

teeth and their successors in the zebrafish and compare the data obtained to those published previously for the mouse.

A summary of zebrafish tooth development with definition of homologous stages in the mouse Pattern of tooth formation Zebrafish teeth develop on the fifth branchial arches (pharyngeal jaws) in three rostro-caudal rows (ventral, mediodorsal, and dorsal), of which the most ventral (V) row has five tooth positions (1V–5V from rostral to caudal). Our study period (48–168 h postfertilization, hpf) covers the development of the first three teeth forming on the ventral row (positions 3V, 4V, and 5V) and of their first successors (replacement teeth) (Fig. 1). Tooth development starts at

The development of each tooth progresses through five phases that can be identified morphologically (see Huysseune et al. 1998; Van der heyden and Huysseune 2000; Van der heyden et al. 2000) (Fig. 2A). Early morphogenesis (EM) covers tooth initiation and beginning of morphogenesis, and is characterized by a thickening of the pharyngeal epithelium (dental placode), which indicates the site of initiation of the first tooth, 4V1 (the only tooth to display such a typical dental placode). Late morphogenesis (LM) is characterized by invagination of the dental epithelium into the underlying mesenchyme, and progressive formation of an asymmetrical bellshaped enamel organ. During early cytodifferentiation (ED), pre-ameloblasts and pre-odontoblasts start to differentiate. Late cytodifferentiation (LD) begins when the ameloblasts and the odontoblasts have fully differentiated, coinciding with the first deposit of the tooth matrices (enameloid and dentin, respectively). Finally, during attachment (not shown), matrix deposition at the tooth base results in ankylosis to the fifth branchial arch. The tooth erupts and becomes functional (cf. Huysseune and Sire 2004). Development is always more advanced at the tip of the tooth than at its base.

Homologous stages in the mouse Six stages are recognized in molar tooth development: dental lamina (initiation), bud, cap, early bell, late bell, and eruption (Zhao et al. 2000). Homologous developmental phases were defined in the two species as follows (Fig. 2B): dental lamina and bud in mice 5 EM in zebrafish; cap 5 LM; early bell 5 ED; late bell 5 LD; attachment 5 eruption. Given the differences in attachment between both species (periodontal ligament vs. ankylosis), we do not report expressions during this period. The main differences between mouse and zebrafish teeth are related to shape (molars in mice vs. canine-like teeth in zebrafish), size and thus number of cells involved (zebrafish first-generation teeth are about one-fifth the size of a mouse molar tooth germ), and structure (the hard tissues are to a certain extent different between zebrafish and tetrapods/mouse: enameloid vs. enamel, atubular dentin at least in first-generation teeth,...), more than to changes in the basic developmental processes.

Borday-Birraux et al.

Dlx expression in zebra¢sh teeth

133

MATERIAL AND METHODS

RESULTS

Animals

In whole-mount hybridized (WM) larvae, teeth showing a signal (Fig. 3) can readily be identified by referring to the timetable of development (Fig. 1). In a ventral view, developing teeth are seen in symmetrical loci, at both sides of the midline, and posterior to the fifth branchial arch (Fig. 3, A and B). Teeth 3V1 and 5V1 develop on both sides of 4V1, with 3V1 closer to the midline and 5V1 more lateral. In a lateral view, developing teeth are identified as round spots located posterior to the fifth branchial arch (Fig. 3, C and D). Accurate data on developmental phase and labeled tissue component(s) can only be obtained from transverse sections through the pharyngeal region, using the toluidine blue stained sections as a reference (Fig. 4, B, H, M, N, and T). Six out of the eight Dlx genes are expressed during morphogenesis and/or cytodifferentiation phases of the first developing teeth (4V1, 3V1, and 5V1) and only some of them during development of the first replacement tooth 4V2. In 144- and 168-hpf specimens, no expression was detected in the developing replacement teeth 3V2 and 5V2. A description of the expression pattern of these genes, based both on WM material, cryosections and on numerous serial sections, is presented below and some examples are illustrated in Figs. 3 and 4.

Zebrafish (D. rerio) embryos originated from our laboratories (Paris, Ghent and Boulder), and were obtained from natural crosses, maintained at 28.51C under standard conditions (Westerfield 1995). Some specimens used for whole-mount in situ hybridization (ISH) were incubated in 1-phenyl-2-thiourea (0.003%) to prevent pigmentation. A series of embryos was fixed from 48 to 120 hpf, for wholemount ISH and from 48 to 168 hpf for ISH on cryosections. Control specimens from the same ontogenetic stages were immersed in a mixture of glutaraldehyde–paraformaldehyde, dehydrated and embedded in epon 812 for standard histology (for more details, see Huysseune et al. 1998).

Whole-mount ISH and sectioning For each stage, the embryos were fixed overnight at 41C in a phosphate-buffered saline (PBS) solution containing 4% paraformaldehyde (PFA), and stored at 201C in methanol. Wholemount ISH was performed as described by Thisse and Thisse (1998) or Jackman et al. (2004).

Probes We-used anti-sense RNA digoxigenin-labeled probes transcribed from cDNA fragments as previously described: dlx2a, dlx3b, and dlx5a (Akimenko et al. 1994); dlx1a, dlx2b, dlx4a, dlx4b, and dlx6a (Ellies et al. 1997). Alternate DNA fragments were used as templates for the synthesis of dlx2a, dlx3b, dlx4b, and dlx5a probes: nucleotides 144–952, 737–1454, 243–1246, and 173–754, respectively (Genbank Accession No: NM_131311, X65060, U67843, and NM_131306).

Whole-mount analysis After hybridization the embryos were postfixed in 4% PFA. Some of them were cleared in glycerol and photographed with Nomarski differential interference contrast (DIC) illumination after removal of the yolk sac. In some cases, handmade cross-sections were performed at the level of the fifth branchial arch.

Transverse sectioning Whole-mount hybridized embryos were dehydrated through a graded series of ethanol, embedded in epon 812, and serially sectioned at 5 mm, from the snout tip backwards, using a diamond knife. Alternatively, specimens were embedded in glycol methacylate (JB-4) and serially sectioned at 2–4-mm thickness. The sections were mounted unstained, and observed and photographed with DIC illumination. Epon-embedded control specimens were serially sectioned at 1 mm, and stained with toluidine blue.

ISH on cryosections Methanol-stored embryos were rehydrated (methanol/PBS series), embedded in cryomount and cryosectioned at 16 mm. The sections were placed on Superfrost/Plus slides coated with poly-L-lysine hydrobromide and hybridized according to Stra¨hle et al. (1994). Hybridization (probe concentration: 500 ng/ml) occurred at 651C overnight. The sections were postfixed in 4% PFA, mounted in aquamount, and observed with DIC.

dlx1a and dlx6a In none of the developmental stages were dlx1a and dlx6a transcripts detected in the dentigerous area (Fig. 3, E and P, respectively), although the expression of these two genes was clearly observed elsewhere in the head.

dlx2a In WM larvae, dlx2a transcripts were observed in the tooth region at 48 hpf and had disappeared at 96 hpf. At 52–56 hpf, the signal appeared as two symmetrical spots on each side of the midline (Fig. 3F), corresponding to the first developing teeth, 4V1. dlx2a expression was never detected in 3V1, 5V1 nor in replacement tooth territories. Sections showed dlx2a transcripts in both the epithelium and mesenchyme of 4V1, during morphogenesis (Fig. 4F) and differentiation phases. During ED, transcripts were seen both in the inner dental epithelial (IDE) cells (pre-ameloblasts) and in the adjacent mesenchymal cells (pre-odontoblasts). During LD, dlx2a expression was downregulated in the upper region of the tooth, whereas it became weak and restricted to the mesenchyme at the tooth base, a region in which odontoblasts were still differentiating (not shown).

dlx2b In WM larvae, dlx2b expression was observed during the development of the first teeth 4V1, 3V1, and 5V1 (Fig. 3, G

134

EVOLUTION & DEVELOPMENT

Vol. 8, No. 2, March^April 2006

Fig. 3. Whole-mount in situ hybridizations for Dlx genes in zebrafish. Pharyngeal region. Anterior is to the left. (A–D), Ventral (A, B) and lateral (C, D) views showing tooth location in 96 hours postfertilization larvae hybridized for dlx4a. (B and D) Interpretative drawings of A and C, respectively. The first tooth to develop (4V1) is located posterior and dorsal to the fifth branchial arch. Other developing teeth (3V1, 5V1, and 4V2) are out of focus. At this stage, 4V1 has attached to the bone support, and dlx4a expression is localized to the tooth base. (E–P) Selected examples of the expression pattern of the eight Dlx genes (ventral views). (E) dlx1a expression is negative in the tooth region (arrows), whereas it is positive in the forebrain and in the mandibular, hyoid and branchial arches I–V. (F) The two labeled cell clusters (arrows) correspond to the first developing tooth, 4V1. dlx2a is also expressed in the developing branchial arches I–V and in the forebrain. (G) 4V1 strongly expresses dlx2b, whereas transcripts are not detected in the branchial arches I–V. (H) The developing teeth 3V1 and 5V1 are labeled by the dlx2b probe. (I) dlx3b is expressed in the early developing 4V1. (J) Developing 3V1 and 5V1 express dlx3b, whereas the expression is maintained in 4V1 (out of focus). dlx3b is also expressed in the branchial arches, except in the fifth one. (K) dlx4a is expressed in the developing 4V1. (L) dlx4a expression is stronger in a more advanced stage of 4V1, but transcripts are not detected in the developing 3V1 and 5V1. This gene is also strongly expressed in branchial arches I–IV, whereas the fifth one shows a weak expression only. (M) A dlx4b signal is observed in the early developing 4V1 tooth germs, but with blurred contours, as in developing branchial arches I–IV. (N) A clear dlx5a expression is seen in the developing 4V1. (O) The dlx5a signal is stronger in late differentiating 4V1. In contrast, in the developing 3V1 and 5V1 the dlx5a signal is too weak to be observed in whole-mounts. The mandibular, hyoid, and branchial arches I–IV are labeled, as well as the forebrain. (P) dlx6a expression is negative in the tooth region (arrows), but the forebrain, and the mandibular, hyoid and branchial arches I–IV are labeled. I–V, branchial arch numbers; ot, otic vesicle; ov, optic vesicle. Scale bars, 100 mm.

and H), as well as in the first replacement tooth, 4V2. At 48 hpf, dlx2b transcription was activated in 4V1 and the signal remained strong until 72 hpf, when two supplementary

bilateral cell clusters, corresponding to 3V1 and 5V1, started to express this gene. At 96 hpf, dlx2b expression became weak in 4V1 and was difficult to observe in WM specimens,

Borday-Birraux et al. whereas 3V1 and 5V1 were strongly labeled (Fig. 3H). In larvae older than 96 hpf, 4V2 started to express dlx2b, but the signal was weak. Sections revealed that dlx2b expression covered all developmental phases of 4V1, 3V1, 5V1, and 4V2, from EM to late differentiation. During EM, dlx2b expression was located in the dental epithelium (Fig. 4, A and S). During LM, the expression extended to the mesenchyme compartment. During early and LD, dlx2b transcripts were localized both within the pre-ameloblasts (IDE) and the pre-odontoblasts (Fig. 4, J, R, and V), but no signal was observed at the tooth base.

Dlx expression in zebra¢sh teeth

135

detected in a region in which 4V1 develops (Fig. 3M). In older stages, because of the strong staining in the pharyngeal region we were not able to detect dlx4b expression in the region corresponding to other developing teeth. Sections showed a clear dlx4b expression in all developing phases of 4V1, 3V1, and 5V1, but not in the replacement teeth. During 4V1 morphogenesis, dlx4b was expressed both in the dental epithelium and in the mesenchyme, where expression was observed throughout the lining of the pharyngeal epithelium. During cytodifferentiation, the signal became restricted to the IDE. In developing 3V1 and 5V1, dlx4b transcripts were similarly restricted to the IDE after morphogenesis (Fig. 4P).

dlx3b In WM larvae, dlx3b expression was observed during the development of 4V1, 3V1, and 5V1, but not of the replacement teeth. In 4V1, the expression was detected as early as 48 hpf and persisted until 97 hpf (Fig. 3I). At 72 hpf, the developing 3V1 and 5V1 started to express dlx3b (Fig. 3J) and remained labeled until 120 hpf, but with a weak signal. Sections showed dlx3b expression in the epithelial compartment during all developmental phases of teeth 4V1, 3V1, and 5V1 (Fig. 4, D, K, L, O) with, however, a transient expression in the mesenchyme during LM (Fig. 4E). As in the case of WM larvae, no expression was seen in sections of replacement teeth.

dlx4a In WM larvae, dlx4a expression was observed exclusively during the development of 4V1 and of its successor, 4V2. In 4V1 the expression was weak at 56 and 77 hpf (Fig. 3, K and L, respectively) and was stronger in older stages (Fig. 3, A and C), in which 4V2 also started to express this gene, but weakly (not shown). Sections showed that dlx4a was expressed in 4V1, from LM to LD. During LM, dlx4a transcripts were localized both in the epithelium and the mesenchyme. During ED the signal increased in the dental papilla (pre-odontoblasts), whereas it weakened in the enamel organ facing these cells (Fig. 4I). During LD, the expression was still stronger in the mesenchyme (differentiating odontoblasts) than in the facing IDE, where ameloblasts were already differentiated. Finally, dlx4a expression was downregulated at the tooth tip, and restricted to the tooth base, essentially in the mesenchyme. The expression disappeared during attachment. A weak expression was also observed during development of the replacement tooth 4V2 (Fig. 4U).

dlx4b In WM larvae, the expression pattern of dlx4b was difficult to assess as a strong labeling of the mesenchyme in the pharyngeal region masked the signal (Fig. 3M). However, in some 56-hpf specimens, a symmetric, weak and diffuse signal was

dlx5a dlx5a expression was activated in the developing 4V1 at 48 hpf. In 60 hpf WM larvae, a weak dlx5a expression was observed in 4V1 (Fig. 3N). The signal was intensified in 72 hpf WM larvae (Fig. 3O) and was still visible in 96-hpf specimens, but was restricted to the tooth base. A weak expression, that is, hardly visible in WM larvae, was observed in the developing 3V1 and 5V1 (Fig. 3O), but dlx5a transcripts were never detected in the replacement teeth. Sections showed that dlx5a was expressed during all developmental phases of 4V1. During EM, dlx5a expression was restricted to the dental epithelium (Fig. 4C). From LM onwards, it was detected both in the IDE and in the facing mesenchyme (Fig. 4G). During LD, dlx5a transcripts were detected at the tooth base, a region in which the ameloblasts and odontoblasts were still differentiating. The signals were strong in the mesenchyme, at the level of the attachment region, both in the odontoblasts and in the osteoblasts lining the bone surface (not shown). In 3V1 and 5V1 a weak expression was observed from LM to LD in the IDE and in the mesenchyme (Fig. 4Q).

DISCUSSION

dlx expression during tooth development in the zebrafish The expression patterns of the eight genes of the Dlx family during the various phases of tooth development in the zebrafish are summarized in Table 2 and Fig. 5A, and commented on below. Two Dlx genes (dlx1a and dlx6a) are not expressed in teeth at any stage, whereas the transcription of the six other genes, dlx2a, dlx2b, dlx3b/dlx4b, dlx4a, and dlx5a, is initiated during the development of the first tooth (4V1). The expression of these genes is maintained from EM to late differentiation of 4V1, and four of them (dlx2b, dlx3b/dlx4b, and dlx5a) are re-expressed in a similar pattern (albeit weaker for dlx5a) during the odontogenetic phases of the two subsequent first-generation teeth, 3V1 and 5V1. The epithelial and me-

136

EVOLUTION & DEVELOPMENT

Vol. 8, No. 2, March^April 2006

senchymal expression domains suggest that dlx2b and dlx5a could be involved in the genetic pathway leading to the differentiation of both ameloblasts and odontoblasts, whereas the linked pair dlx3b/dlx4b, whose expression is restricted to the dental epithelium during the early and and late differentiation stage, are more likely to be involved in ameloblast differentiation. The two other Dlx genes (dlx2a and dlx4a) are downregulated when 4V1 has been completed and are not re-activated during the development of the subsequent first-generation teeth. However, dlx4a is re-expressed during 4V2 development. The more recently duplicated paralogs dlx2a and dlx2b, and dlx4a and dlx4b, show differences in their temporal expression: dlx2b and dlx4b are expressed during development of the first-generation teeth (4V1, 3V1, 5V1), whereas dlx2a and dlx4a are expressed during development of

the first tooth only. This downregulation of dlx2a and dlx4a could be explained by possible redundant functions with their recently formed paralogs. dlx2a being the only gene of the family to be expressed only in 4V1, this could support dlx2a as playing a role in patterning the first generation teeth. We use ‘‘patterning’’ to refer to the putative signal rendering dental competence to the pharyngeal epithelium in the region where the first-generation teeth will be formed. This ‘‘patterning’’ is similar to the model of tooth primordia initiation proposed by Smith (2003), that is, the initiation of the dental pattern from a single dental determinant (primordial toothsignaling center) with a subsequent progression in one or several directions. Only two Dlx genes (dlx2b, dlx4a) are expressed in the first replacement tooth 4V2 (with fairly similar patterns). This

Borday-Birraux et al.

Dlx expression in zebra¢sh teeth

Table 2. Expression pattern of the eight Dlx genes during development of the first-generation teeth (4V1, 3V1, and 5V1) and of the first replacement tooth (4V2) in zebrafish Developmental Phases Gene Tooth dlx1a None dlx2a 4V1 dlx2b 4V1, 3V1 5V1, 4V2 dlx3b 4V1, 3V1, 5V1 dlx4a 4V1, 4V2 dlx4b 4V1, 3V1, 5V1 dlx5a 4V1, 3V1, 5V1

Compartment EM LM ED Epithelium Mesenchyme Epithelium Mesenchyme Epithelium Mesenchyme Epithelium Mesenchyme Epithelium Mesenchyme Epithelium Mesenchyme

1 1 1

11 11 11 11 11 11 1 1 1 1 1 1 1 11 11 1

11 11 11 11 11

LD

1b 11 11 11

1 1 11 11b 1 1 11 11b 1 1b

dlx6a None Two compartments (epithelium and mesenchyme) and four developmental phases (early and late morphogenesis (EM, LM), early and late cytodifferentiation (ED, LD)) have been distinguished. Note that neither dlx1a nor dlx6a are expressed in any compartment during tooth development in the zebrafish. , no signal; 1, weak signal; 11, strong signal; b, signal at the tooth base.

suggests that differences exist between the genetic control of 4V1 development, during which six Dlx genes are expressed, and of its successor, where only two Dlx genes are expressed.

137

The smaller number of genes expressed in replacement teeth is intriguing. Considering the fact that, in humans, mutations of the MSX1 and PAX9 genes affect only the replacement teeth but not the deciduous teeth (Vastardis et al. 1996; Stockton et al. 2000), one might be tempted to hypothesize that the expression of fewer redundant paralogs buffers replacement teeth less against mutations than their predecessors. Results obtained from in vitro cultures support the notion that initiation of zebrafish replacement teeth may be under a different control than that of first-generation teeth (Huysseune and Thesleff 2004; Van der heyden et al. 2005). It has been proposed that this difference relates to the possibility that a stem cell population is established only at the start of the first replacement event (Huysseune and Thesleff 2004). On the other hand, dlx2b could play a role in tooth primordium initiation of the replacement teeth, similar to the role proposed for dlx2a in first-generation teeth. The lack of Dlx signal during development of the two other replacement teeth 3V2 and 5V2 could be related to the late development of these teeth (at the 6th day) and, therefore, to the detection limits of our ISH techniques. Although several Dlx genes are expressed late in tooth development in relation to the presence of still differentiating ameloblasts and odontoblasts along the tooth shaft, only dlx5a is clearly and strongly expressed at the tooth base during attachment. Possibly, dlx5a is involved in the control of cell differentiation in the region where the tooth base will ankylose to the bone support, the now ossified ceratobranchial V (fifth branchial arch). The matrix of this perichondral bone is synthesized by a layer of osteoblasts, which also express dlx5a (L. Verreijdt et al., in prep.). Thus, the same regulatory gene, dlx5a, is expressed simultaneously in various cell

Fig. 4. Selected examples of the expression pattern of Dlx genes during zebrafish tooth morphogenesis and differentiation. Transverse sections passing through the pharyngeal region at the level of the fifth branchial arch. Some sections of whole-mount hybridized larvae are completed with an interpretative drawing (mirror image). Reference sections are shown for each phase of tooth development, which is indicated in the upper right corner: EM, early morphogenesis; LM, late morphogenesis; ED, early cytodifferentiation; LD, late cytodifferentiation. (A–K) Expression pattern of Dlx genes during 4V1 development. (A–D) Early morphogenesis phase. (A) 48 hours postfertilization (hpf); (B) 48 hpf: control toluidine blue-stained specimen. Tooth 4V1 appears as a placode (arrow). (C) 48 hpf; (D) 52 hpf. dlx2b, dlx3b, and dlx5a are both expressed in the epithelium compartment. (E–F) Late morphogenesis phase. Expression pattern at 56 and 52 hpf, respectively. dlx2a and dlx3b are expressed both in the mesenchyme and in the epithelium, which has started to invaginate into the mesenchyme. (G–I) Early differentiation phase. (G) 72 hpf, handmade section. (H) 72 hpf, control, toluidine blue-stained specimen. The ameloblasts of tooth 4V1 are differentiating (arrow). The arrowheads point to the regions in which teeth 3V1 and 5V1 are beginning to form (out of focus). (I) 77 hpf, handmade section. dlx5a and dlx4a are expressed in both differentiating ameloblasts and odontoblasts. (J–K) Late differentiation phase. At 74 and 96 hpf, respectively, dental matrix has accumulated (arrowheads). dlx2b is expressed both in the ameloblasts and odontoblasts, whereas dlx3b expression is restricted to the ameloblasts. (L–R) Expression pattern of Dlx genes in the developing teeth 3V1 and 5V1. (L) Early morphogenesis phase. 72 hpf. dlx3b is expressed in the epithelium. (M–Q) Early differentiation phase. (M–N) 80 hpf. Control, toluidine blue-stained specimens showing 3V1 (M) and 5V1 (N). (O–Q) 96 hpf. dlx5a is expressed in both differentiating ameloblasts and odontoblasts, whereas dlx3b and dlx4b expression is restricted to the differentiating ameloblasts. Note that these three genes are expressed during attachment of 4V1. (R) Late differentiation phase. 96 hpf, dlx2b expression in both the ameloblasts and odontoblasts of 3V1. Note the expression of dlx2b during attachment of 4V1 in the ameloblasts at the tip of the tooth. (S–V) Expression of Dlx genes in the developing replacement tooth, 4V2. dlx2b is strongly expressed in both ameloblasts and odontoblasts, whereas a weak dlx4a expression is restricted to the odontoblasts. (S) Early morphogenesis phase. 96 hpf. (T–U) Early differentiation phase. (T) 120 hpf control, toluidine bluestained specimen. (U) 144 hpf. (V) Late differentiation phase, 120 hpf. c, notochord; h, hindbrain; phe, pharyngeal epithelium; phm, pharyngeal mesenchyme; ot, otic vesicle; y, yolk; Vth, fifth branchial arch. Scale bars: A, B, G–V 5 50 mm; C–F 5 25 mm.

138

EVOLUTION & DEVELOPMENT

Vol. 8, No. 2, March^April 2006

Fig. 5. Comparison of Dlx gene expression patterns during the various development phases of tooth 4V1 in the zebrafish (A, eight Dlx genes) and of the molar in the mouse (B, six Dlx genes). In the mouse, the two first developmental stages (initiation and bud) have been grouped and considered comparable with the developmental phase of early morphogenesis in the zebrafish. Orthologous genes are linked by dotted lines. Genes of the same linked pair are connected by a double arrow. (B) From Zhao et al. (2000). Red, epithelial expression; blue, mesenchymal expression. EM, early morphogenesis; LM, late morphogenesis; ED, early cytodifferentiation; LD, late cytodifferentiation; La, labial side; Li, lingual side.

populations and may aid in achieving an efficient anchoring of the tooth base to the bone support.

Dlx genes and redundancy In general, the co-expression of several genes belonging to a single regulatory gene family suggests possible functional redundancy. This has already been reported for Hox (Horan et al. 1995; Favier et al. 1996), Msx (Catron et al. 1996; Zhang et al. 2003), and particularly Dlx genes in mice (Qiu et al. 1997; Depew et al. 1999; Zerucha and Ekker 2000). In the zebrafish, the overlap of expression patterns of six Dlx genes both in time (similar phases of development) and in space (similar tissue compartments) during the formation of the first tooth, 4V1, could indeed suggest redundancy (Table 2, Fig. 5A). For instance, during LM of 4V1, the six Dlx genes are expressed in the dental epithelium and in the dental papilla. However, such a co-expression could also be related to the

rapid development of 4V1. Indeed, this tooth is initiated, develops and becomes functional within a 24-h period. It is therefore difficult to distinguish whether the overlap is total or only partial or complementary. In addition, the tooth is very small, and the enamel organ and dental papilla are made up of a small number of cells only. One can question whether up and downregulation can be detected in such circumstances. The eight zebrafish Dlx genes have different degrees of relatedness based on their appearance in several successive episodes in chordate evolution (Stock et al. 1996; Neidert et al. 2001; Stock 2005). The earliest divergence separates dlx1a, dlx4a, dlx4b, and dlx6a from dlx2a, dlx2b, dlx3b, and dlx5a. Linked pairs of genes consist of members of both of these two clades. Relationships within the clades remain unclear, except for the close relationship of dlx2a to dlx2b and of dlx4a to dlx4b, reflecting a relatively recent genome duplication in rayfinned fishes. It can be asked whether co-expression of Dlx

Borday-Birraux et al. genes (and possible functional redundancy) is most characteristic of (i) genes within the same major clade, (ii) genes on the same chromosome (distantly related at the sequence level, but with the potential for shared regulatory elements), or (iii) recent ray-finned fish duplicates. Interestingly, the only genes with the same temporal and spatial patterns of expression in tooth 4V1 are dlx2b and dlx5a (Fig. 5), unlinked members of the same major clade. Differences in expression in 4V1 exist within both pairs of recently duplicated genes (Dlx2 and Dlx4 paralogs), occurring at one and three of the four stages examined, respectively. In the mouse, the two members of the convergently transcribed Dlx pairs (Dlx1/Dlx2, Dlx3/Dlx4, and Dlx5/Dlx6) are expressed in roughly similar patterns in, for example, forebrain, branchial arches or sensory placodes, and this generally is the case in the homologous organs of the zebrafish as well (Zerucha and Ekker 2000; Zerucha et al. 2000). In contrast, during zebrafish tooth development, only linked dlx3b/dlx4b genes comply with this rule: they are coexpressed, except during EM where dlx3b expression is not detected in the mesenchyme (Fig. 5A). This suggests possible redundancy for dlx3b and dlx4b during tooth development, as already reported for the otic placode and vesicle (Solomon and Fritz 2002). Although mice Dlx3 mutants die at birth, tooth defects were not reported, which indicates possible redundancy (Morasso et al. 1999). In contrast, in humans a DLX3 mutation leads to a syndrome (tricho-dento-osseous), in which teeth are affected (Price et al. 1998a, b). In the other linked gene pairs, one member of the pair, dlx1a and dlx6a, is never transcribed, in contrast to the other member, dlx2a and dlx5a, respectively, which shows strong expression. However, redundancy is often only valid at first glance, but not when expression patterns are studied in detail (e.g., Quint et al. 2000; Robledo et al. 2002; Solomon and Fritz 2002). In the mouse for instance, the pairs Dlx1/Dlx2 and Dlx5/Dlx6 have different expression patterns during tooth development, suggesting specific functions as, for example, in the dental sac (Fig. 5B). However, using targeted mutation in the mouse, redundancy has been demonstrated for these genes. Neither Dlx1 nor Dlx2 mutant mice have a clear defect in tooth development (Qiu et al. 1997), but the double mutant Dlx1/Dlx2 / fails to develop maxillary molar teeth, because of a defect in the mesenchyme (Thomas et al. 1997). Similarly, targeted mutation of Dlx6 or of Dlx5 has no important effect on the mandible and tooth development (Acampora et al. 1999; Depew et al. 1999), whereas double-mutant mice Dlx5/Dlx6 / fail to develop the mandible, but no data are available for possible tooth defects (Robledo et al. 2002). Therefore, in the mouse there is both redundancy and specific function for the two genes of a linked pair. In the latter case for instance, in the mouse one member of the pair has a specific function in the dental sac, but not the other member: Dlx1 but not Dlx2, Dlx6 but not Dlx5. Unfortunately, the knockout results do not rule out these

Dlx expression in zebra¢sh teeth

139

different functions between members of a linked pair because the unlinked Dlx1 and Dlx6 could be redundant in the dental sac.

A comparison of Dlx gene expression during tooth development in the zebrafish and the mouse Above, we have highlighted some differences in the expression of various Dlx genes in the zebrafish between the first tooth to be formed, 4V1, the next two first-generation teeth, 3V1 and 5V1, and the replacement tooth 4V2. This suggests that different tooth positions (4V1 vs. 3V1 and 5V1), as well as different tooth generations (4V1 vs. 4V2), do not share the same molecular pathway and need to be studied in a diphyodont species in order to check whether differences exist in Dlx gene expression patterns. In the absence of such data, we have compared the expression patterns of the Dlx genes in tooth 4V1 in the zebrafish (Fig. 5A), with those in the mouse first molar, as reported by Zhao et al. (2000) (Fig. 5B). The most striking differences are : 1. Two dlx genes involved in murine tooth development are not expressed during zebrafish tooth development dlx1a and dlx6a transcripts were never detected during zebrafish tooth development in spite of the numerous specimens examined. It is unlikely that this negative result could have resulted from problems with the probes, as both genes were strongly expressed in other regions of the head and other Dlx gene transcripts were detected during similar phases of tooth development. In molar tooth development of the mouse, Dlx1 transcripts have been detected in a layer of mesenchymal cells bordering the dental papilla ( 5 the dental sac) from LM to LD (Zhao et al. 2000). Surprisingly, this pattern of expression is similar to that described for Dlx6, whose transcripts have been detected in the same odontogenetic phases, and later also in differentiating odontoblasts (Fig. 5B). The absence of dlx1a and dlx6a expression during zebrafish tooth development could, therefore, be related to the absence of such a dental sac. Three evolutionary hypotheses can be proposed to explain differences between the zebrafish and mouse in Dlx1 and Dlx6 ortholog expression. One is that Dlx1 and Dlx6 involvement in mammalian tooth development is recent, meaning both genes were recruited in a mammalian ancestor, possibly during the polyphyodont–diphyodont transition, which was accompanied by drastic changes in the mode of tooth attachment. Alternatively, the genes ancestral to Dlx1/dlx1a and Dlx6/dlx6a played a role in tooth development in an early osteichthyan, but dlx1a and dlx6a lost this function during evolution of the actinopterygian lineage leading to the zebrafish. This loss could be explained either because their specific function was taken over by the other pair member, dlx2a and dlx5a, or by another gene, or because the developmental fea-

140

EVOLUTION & DEVELOPMENT

Vol. 8, No. 2, March^April 2006

tures for which their activation was necessary, that is, the formation of the dental sac, disappeared. Finally, Dlx1 and Dlx6 ortholog expression may have characterized oral but not pharyngeal teeth in early osteichthyans. Additional data are required in other lineages to test these hypotheses. 2. Tooth development in the zebrafish and the mouse show both similarities and distinctive differences in the expression of Dlx genes Together, the expression patterns of the six Dlx genes involved in zebrafish tooth development cover both the period and domains of expression of the four orthologous Dlx genes in the mouse. However, none of the genes shows a completely identical expression pattern when considering the four developmental phases and the two compartments (epithelium and mesenchyme). In both tissues, the transcripts were nevertheless detected over a longer period of development in the zebrafish than in the mouse, where they were restricted to specific phases. These differences could be related to the speed of tooth development in the zebrafish (each phase taking only a few hours), to the simpler (and likely less derived) dentition (Sire et al. 2002), or to both. In addition, the genome duplication, which has occurred in the teleost lineage (Amores et al. 1998) could have facilitated a redistribution of Dlx gene expression and function during odontogenesis. Remarkably, more Dlx genes are detected in the epithelium in the zebrafish than in the mouse, even when genome duplication is taken into account. For example, dlx4a and dlx4b show a strong epithelial expression, whereas mouse Dlx4 is activated in the dental organ during LD, but not in the inner dental epithelium. Similarly, zebrafish dlx5a is expressed in the epithelium and in the mesenchyme during all phases of odontogenesis while Dlx5 transcripts are reported in the mesenchyme only. If one considers EM exclusively, several zebrafish Dlx genes are expressed in a way reminiscent of the situation in the mouse: dlx2a1dlx2b 5 Dlx2; dlx3b 5 Dlx3; dlx4a 5 Dlx4 (Zhao et al. 2000). During LM, six Dlx genes are expressed in the mesenchyme, as is the case in the mouse. These data suggest a functional conservation of several Dlx genes in both lineages for the early stage of development. The different expression patterns observed for several Dlx genes during the cytodifferentiation phase are likely to be related to the complex differentiation processes associated with cusp formation in the mouse, and notably with the formation of the enamel knot as a signaling center (Vaahtokari et al. 1996; Laurikkala et al. 2003; Luukko et al. 2003). Alternatively, the different locations of the teeth (pharyngeal vs. oral) could account for the difference in gene expression during tooth development. The loss of teeth in the oral cavity is assumed to have occurred in a cypriniform ancestor. Studies of Dlx genes in species retaining both oral and pharyngeal teeth will be required to determine which of the expression differences we have identified between zebrafish and

mouse teeth are characteristic of differences between oral and pharyngeal teeth. Acknowledgments We gratefully acknowledge Marc Ekker and Marie-Andre´e Akimenko (Loeb Institute, Ottawa) for supplying some of the plasmids, Ce´line Goveovich and Franc¸oise Allizard for technical help, and the CNRS (UMR 7138, and PICS 1890) (J. Y. S.), (UPR 9034) (V. B. B.), the ‘‘Bijzonder Onderzoeksfonds’’ of Ghent University (011V1298) (A. H.), and NSF (IBN-0092487) (D. W. S.) for grants.

REFERENCES Acampora, D., et al. 1999. Cranofacial, vestibular and bone defects in mice lacking the Distal-less-related gene Dlx5. Development 126: 3795–3809. Akimenko, M. A., Ekker, M., Wegner, J., Lin, W., and Westerfield, M. 1994. Combinatorial expression of three zebrafish genes related to distal-less: part of a homeobox gene code for the head. J. Neurosci. 14: 3475–3486. Amores, A., et al. 1998. Zebrafish hox clusters and vertebrate genome evolution. Science 282: 1711–1714. Bei, M., and Maas, R. 1998. FGFs and BMP4 induce both Msx1-independent and Msx1-dependent signaling pathways in early tooth development. Development 125: 4325–4333. Catron, K. M., Wang, H., Hu, G., Shen, M. M., and Abate-Shen, C. 1996. Comparison of MSX-1 and MSX-2 suggests a molecular basis for functional redundancy. Mech. Dev. 55: 185–199. Depew, M. J., et al. 1999. Dlx5 regulates regional development of the branchial arches and sensory capsules. Development 126: 3831–3846. Ekker, M., Akimenko, M.-A., Bremiller, R., and Westerfield, M. 1992. Regional expression of three homeobox transcripts in the inner ear of zebrafish embryos. Neuron 9: 27–35. Ellies, D. L., et al. (1997). Relationship between the genomic organization and the overlapping embryonic expression patterns of the zebrafish dlx genes. Genomics 45: 580–590. Favier, B., et al. 1996. Functional cooperation between the non-paralogous genes Hoxa-10 and Hoxd-11 in the developing forelimb and axial skeleton. Development 122: 449–460. Fraser, G. J., Graham, A., and Smith, M. M. 2004. Conserved deployment of genes during odontogenesis across osteichthyans. Proc. Roy. Soc. London B 271: 2311–2317. Horan, G. S., et al. 1995. Compound mutants for the paralogous hoxa-4, hoxb-4, and hoxd-4 genes show more complete homeotic transformations and a dose-dependent increase in the number of vertebrae transformed. Genes Dev. 13: 1667–1677. Huysseune, A., and Sire, J.-Y. 2004. The role of epithelial remodelling in tooth eruption in larval zebrafish. Cell Tissue Res. 315: 85–95. Huysseune, A., and Sire, J. Y. 1998. Evolution of patterns and processes in teeth and tooth-related tissues in non-mammalian vertebrates. Eur. J. Oral Sci. 106: 437–481. Huysseune, A., and Thesleff, I. 2004. Continuous tooth replacement: the possible involvement of epithelial stem cells. BioEssays 26: 665–671. Huysseune, A., Van der heyden, C., and Sire, J.-Y. 1998. Early development of the zebrafish (Danio rerio) pharyngeal dentition (Teleostei, Cyprinidae). Anat. Embryol. 198: 289–305. Hsiao, C.-D., Tsai, W.-Y., and Tsai, H.-J. 2002. Isolation and expression of two zebrafish homologs of parvalbumin gene related to chicken CPV3 and mammalian oncomodulin. Gene Express. Patterns 2: 163–168. Jackman, W. R., Draper, B. D., and Stock, D. W. 2004. Fgf signaling is required for zebrafish tooth development. Dev. Biol. 274: 139–157. Kera¨nen, S. V. E., Aberg, T., Kettunen, P., Thesleff, I., and Jernvall, J. 1998. Association of developmental regulatory genes with the development of different molar tooth shapes in two species of rodents. Dev. Genes Evol. 208: 477–486. Laurenti, P., Thaeron, C., Allizard, F., Huysseune, A., and Sire, J.-Y. 2004. Cellular expression of eve1 suggests its requirement for the differentiation of the ameloblasts and for the initiation and morphogenesis of the first tooth in the zebrafish (Danio rerio). Dev. Dyn. 230: 727–733.

Borday-Birraux et al. Laurikkala, J., Kassai, Y., Pakkkasja¨rvi, L., Thesleff, I., and Itoh, N. 2003. Identification of a secreted BMP antagonist, ectodin, integrating BMP, FGF, and SHH signals from the tooth enamel knot. Dev. Biol. 264: 91–105. Luukko, K., Loes, S., Furmanek, T., Fjeld, K., Kvinnsland, I. H., and Kettunen, P. 2003. Identication of a novel putative signaling center, the tertiary enamel knot in the postnatal mouse molar tooth. Mech. Dev. 120: 270–276. Morasso, M. I., Grinberg, A., Robinson, G., Sargent, T. D., and Mahon, K. A. 1999. Placental failure in mice lacking thte homeobox gene Dlx3. Proc. Natl. Acad. Sci. USA 96: 162–167. Nakamura, S., et al. (1996). Genomic analysis of a new mammalian distalless gene: Dlx7. Genomics 38: 314–324. Neidert, A. H., Virupannavar, V., Hooker, G. W., and Langeland, J. A. 2001. Lamprey Dlx genes and early vertebrate evolution. Proc. Natl. Acad. Sci. USA 98: 1665–1670. Nieminen, P., Pekkanen, M., Aberg, T., and Thesleff, I. 1998. A graphical WWW-database on gene expression in tooth. Eur. J. Oral Sci. 106: 7–11. Payne, T. L., Skobe, Z., and Yelick, P. C. 2001. Regulation of tooth development by the novel type I TGFbeta family member receptor Alk8. J. Dent. Res. 80: 1968–1973. Panganiban, G., and Rubenstein, J. L. R. (2002). Developmental functions of the Distal-less/Dlx homeobox genes. Development 129: 4371–4386. Price, J. A., Bowden, D. W., Wright, J. T., Pettenati, M. J., and Hart, T. C. 1998a. Identification of a mutation in DLX3 associated with trichodento-osseous (TDO) syndrome. Hum. Mol. Genet. 7: 563–569. Price, J. A., Wright, J. T., Kula, K., Bowden, D. W., and Hart, T. C. 1998b. A common DLX3 gene mutation is responsible for tricho-dento-osseous (TDO) syndrome in Virginia and North Carolina families. J. Med. Genet. 35: 825–828. Qiu, M., et al. 1997. Role of Dlx homeobox genes in proximodistal patterning of the branchial arches: mutations of Dlx-1, Dlx-2, and Dlx-1 and -2 alter morphogenesis of proximal skeletal and soft tissue structures derived from the first and second arches. Dev. Biol. 185: 165–184. Quint, E., Zerucha, T., and Ekker, M. 2000. Differential expression of orthologous Dlx genes in zebrafish and mice: implications for the evolution of the Dlx homeobox gene family. J. Exp. Zool. 288: 235–241. Robinson, G. W., and Mahon, K. A. 1994. Differential ad overlapping expression domains of Dlx-2 and Dlx-3 suggest distinct roles for Distalless homeobox genes in craniofacial development. Mech. Dev. 48: 199–215. Robledo, R. F., Rajan, L., Li, X., and Lufkin, T. (2002). The Dlx5 and Dlx6 homeobox genes are essential for craniofacial, axial and appendicular skeletal development. Genes Dev. 16: 1089–1101. Rudel, D., and Sommer, R. J. 2003. The evolution of developmental mechanisms. Dev. Biol. 264: 15–37. Scherer, S. W., et al. 1995. Assignment of the human homolog of mouse Dlx3 to chromosome 17q21.3–q22 by analysis of somatic cell hybrids and fluorescence in situ hybridization. Mamm. Genome 6: 310–311. Simeone, A., et al. 1994. Cloning and characterization of two members of the vertebrate Dlx gene family. Proc. Natl. Acad. Sci. USA 91: 2250– 2253. Sire, J.-Y., Davit-Be´al, T., Delgado, S., Van der heyden, C., and Huysseune, A. 2002. The first generation teeth in non-mammalian lineages: evidence for a conserved ancestral character? Microsc. Res. Technol. 59: 408–434. Sire, J.-Y., and Huysseune, A. 2003. Formation of skeletal and dental tissues in fish: a comparative and evolutionary approach. Biol. Rev. 78: 219–249.

Dlx expression in zebra¢sh teeth

141

Smith, M. M. 2003. Vertebrate dentitions at the origin of jaws: when and how pattern evolved. Evol. Dev. 5: 394–413. Solomon, K., and Fritz, A. 2002. Concerted action of two dlx paralogs in sensory placode formation. Development 129: 3127–3136. Stock, D. W. 2005. The Dlx gene complement of the leopard shark, Triakis semifasciata, resembles that of mammals: implications for genomic and morphological evolution of jawed vertebrates. Genetics 104: 807–817. Stock, D. W. 2001. The genetic basis of modularity in the development and evolution of the vertebrate dentition. Philos. Trans. Roy. Soc. London B 356: 1633–1653. Stock, D. W., et al. 1996. The evolution of the vertebrate Dlx gene family. Proc. Natl. Acad. Sci. USA 93: 10858–10863. Stockton, D. W., et al. 2000. Mutation of PAX9 is associated with oligodontia. Nat. Genet. 24: 18–19. Stra¨hle, U., Blader, P., Adam, J., and Ingham, P. W. 1994. A simple and effiicent procedure for non-isotopic in situ hybridization to sectioned material. Trends Genet. 10: 75–76. Thisse, C., and Thisse, B. 1998. High Resolution Whole-Mount In Situ Hybridization. Zebrafish Science Monitor. Vol. 5. University of Oregon Press, Eugene [Zfin database: http://www-igbmc.u-strasbg.fr/zf-info/ zbook/chapt9/9.82.html]. Thomas, B. L., et al. 1997. Role of Dlx-1 and Dlx-2 genes in patterning of the murine dentition. Development 124: 4811–4818. Thomas, B. L., Liu, J. K., Rubenstein, J. L., and Sharpe, P. 2000. Independent regulation of Dlx2 expression in the epithelium and mesenchyme of the first branchial arch. Development 127: 217–224. Vaahtokari, A., Aberg, T., Jernvall, J., Keranen, S., and Thesleff, I. 1996. The enamel knot as a signaling center in the developing mouse tooth. Mech. Dev. 54: 39–43. Van der heyden, C., Allizard, F., Sire, J.-Y., and Huysseune, A. 2005. Tooth development in vitro in two teleost fish, the cichlid Hemichromis bimaculatus and the cyprinid Danio rerio. Cell Tissue Res. 321: 375–389. Van der heyden, C., and Huysseune, A. 2000. Dynamics of tooth formation and replacement in the zebrafish (Danio rerio ) (Teleostei, Cyprinidae). Dev. Dyn. 219: 486–496. Van der heyden, C., Huysseune, A., and Sire, J.-Y. 2000. Development and fine structure of pharyngeal teeth in juvenile zebrafish (Danio rerio) (Teleostei, Cyprinidae). Cell Tissue Res. 302: 205–219. Vastardis, H., et al. 1996. A human MSX1 homeodomain missense mutation causes selective tooth agenesis. Nat. Genet. 13: 417–421. Weiss, K. M., Bollekens, J., Ruddle, F. H., and Takashita, K. 1994. Distalless and other homeobox genes in the development of the dentition. J. Exp. Zool. 270: 273–284. Weiss, K. M., Ruddle, F. H., and Bollekens, J. 1995. Dlx and other homeobox genes in the morphological development of the dentition. Connect. Tissue Res. 32: 35–40. Westerfield, M. 1995. The Zebrafish Book. Guide for the Laboratory Use of Zebrafish (Danio rerio). 3rd Ed. University of Oregon Press, Eugene, OR. Zhang, Z., et al. 2003. Malleal processus brevis is dispensable for normal hearing in mice. Dev Dyn. 227: 69–77. Zerucha, T., and Ekker, M. 2000. Distal-less-related homeobox genes of vertebrates: evolution, function, and regulation. Biochem. Cell Biol. 78: 593–601. Zerucha, T., et al. 2000. A highly conserved enhancer in the Dlx5/Dlx6 intergenic region is the site of cross-regulatory interactions between Dlx genes in the embryonic forebrain. J. Neurosci. 20: 709–721. Zhao, Z., Stock, D. W., Buchanan, A. V., and Weiss, K. M. (2000). Expression of Dlx genes during the development of the murine dentition. Dev. Genes Evol. 210: 270–275.