EVOLUTION & DEVELOPMENT
16:4, 197–206 (2014)
DOI: 10.1111/ede.12083
Comparative gene expression analyses reveal heterochrony for Sox9 expression in the cranial neural crest during marsupial development Yoshio Wakamatsu,a Tadashi Nomura,b Noriko Osumi,a and Kunihiro Suzukic a
Department of Developmental Neuroscience, United Centers for Advanced Research and Translational Medicine (ART), Tohoku University Graduate School of Medicine, Sendai, Miyagi 980‐8575, Japan b Department of Biology, Kyoto Prefectural University of Medicine, Kyoto 603‐8334, Japan c Department of Biology, Nihon University School of Dentistry at Matsudo, Chiba 271‐8587, Japan *Author for correspondence (e‐mail: [email protected])
SUMMARY Compared to placental mammals, marsupials have short gestation period, and their neonates are relatively immature. Despite these features, marsupial neonates must travel from the birth canal to the teat, suckle and digest milk to complete development. Thus, certain organs and tissues of marsupial neonates, such as forelimbs to crawl and jaw elements to suckle, must develop early. Previous reports showed that cranial neural crest (CNC) cells, as the source of ectomesenchyme of jaw elements, are generated significantly early in gray short‐tailed opossum (Monodelphis domestica) compared to other amniote models, such as mouse. In this study, we examined the expression of genes known to be important for neural crest formation, such as BMP2/BMP4 (neural crest inducer), Pax7 (neural border specifier), Snail1 and Sox9/Sox10 (neural crest specifier) in Monodelphis
domestica, and compared the expression patterns with those in mouse, chicken, and gecko embryos. Among those genes, the expression of Sox9 was turned on early and broadly in the premigratory CNC cells, and persisted in the ectomesenchyme of the cranial anlagen in opossum embryos. In contrast, Sox9 expression diminished in the CNC cells of other animals at the early phase of migration. Comparison of the onset of Pax7 and Sox9 expression revealed that Sox9 expression in the prospective CNC was earlier and broader than Pax7 expression in opossum, suggesting that the sequence of border specification and neural crest specification is altered. This study provides the first clue for understanding the molecular basis for the heterochronic development of the CNC cells and jaw elements in marsupials.
INTRODUCTION
their mating has to be visually confirmed), and therefore has been used in a variety of evolutionary developmental studies including the heterochronic development of limbs (Keyte and Smith 2010; Debora and Sears 2010), the brain (Cheung et al. 2009), and the tooth (Moustakas et al. 2011). Completion of genome sequencing will also facilitate the use of M. domestica. In previous studies, Smith and colleagues have shown that cranial neural crest (CNC) cells of M. domestica embryos delaminate, and start forming cranial processes, such as mandibular, maxillary, and hyoid arches, relatively earlier compared to those of eutherian embryos (Vaglia and Smith 2003; see also Smith, 2001 and 2006 for reviews). It is not known, however, how this heterochronic development of the CNC is molecularly and genetically regulated. Neural crest (NC), which was referred to as “the fourth germ layer” specific in vertebrates (Hall 2000, and references therein), gives rise to a wide variety of tissues and cells types, including neurons and glial cells of the peripheral nervous system, pigment cells in the skin, and skeletal and connective tissues in the head (Le Douarin and Kalcheim 1999). Among those, the ectomesenchymal subpopulations specific to the CNC contribute
Marsupial (metatherian) mammals, such as kangaroos and koalas, have significantly short gestation periods compared to placental (eutherian) mammals, and their neonates are, thus, relatively altricial at birth (see Smith 2001 and references therein). Despite their immature state of developmental stage, marsupial neonates must travel form the birth canal to the teat, suckle, and digest milk to complete the rest of their development. Therefore, certain tissues and organs required for such behavior have to develop relatively early during their short embryonic period. Among such organs, craniofacial structures, such as jaws and throat, are important for suckling the milk and breathing, and previous morphological and histological studies by Smith and her colleagues have revealed the remarkably early, “heterochronic” development of the cranial anlagen in the embryo of the gray short‐tailed opossum, Monodelphis domestica (M. domestica, Vaglia and Smith 2003; see also Smith 2001, 2006 for reviews). M. domestica is favored as an experimental laboratory marsupial, because of its relatively small size (between mouse and rat), ease of cultivation in captivity, and availability of embryos at desired stages (although © 2014 Wiley Periodicals, Inc.
197
198
EVOLUTION & DEVELOPMENT
Vol. 16, No. 4, July–August 2014
to generating the morphological diversity of cranium and teeth among vertebrates. There are numerous reports and reviews describing the regulatory mechanisms of the NC induction, epithelial– mesenchymal transition (EMT), migration, and fate specification (For reviews, see Sakai and Wakamatsu 2005; Betancur et al. 2010; Stuhlmiller and García‐Castro 2012; Milet and Monsoro‐Burq 2012; Theveneau and Mayor 2012; McKeown et al. 2013). As for NC induction, several models have been proposed, and the latest is as follows (see also reviews mentioned above). Initially a non‐neural/neural border region in the ectoderm is specified under the influence of Wnt, FGF, and BMP signaling, by inducing the expression of “border specifier” genes such as Pax7 (e.g., Basch et al. 2006). Subsequently, appropriate levels of Wnt and BMP signalings further induce the expression of “neural crest specifier” transcription factor genes, such as Sox9/Sox10, Snail1/Snail12, at the border region (e.g., García‐Castro et al. 2002; Endo et al. 2002; Sakai et al. 2005). These transcription factors in turn control the expression of downstream genes involved in the cell–cell adhesion, cell– extracellular matrix adhesion, cytoskeletal regulation, as well as differentiation (e.g., Cheung et al. 2005; Sakai et al. 2006; Taneyhill et al. 2007; Suzuki et al. 2010; Krispin et al. 2010). In this study, to obtain insights into how the heterochronic development of the marsupial CNC cells is regulated, we examined expression patterns of genes involved in CNC formation in M. domestica embryos, and compared them with those in other amniotes. Among the genes examined, Sox9 was expressed early in a relatively broad area of the ectoderm, likely corresponding to the prospective CNC domain, and the expression persisted longer in the CNC cells during the development of the cranial anlagen, compared to that in mouse, chicken, and gecko. As the function of Sox9 is important for NC formation, EMT, and subsequent differentiation of ectomesenchymal components in the developing CNC cells (Mori‐ Akiyama et al. 2003; Akiyama et al. 2004; Sakai et al. 2006. See also Sakai and Wakamatsu 2005; Milet and Monsoro‐ Burq 2012; Theveneau and Mayor 2012 for reviews), we suggest that the spatio‐temporal alterations of Sox9 expression in the CNC cells may be the key for the heterochronic development of cranial elements of marsupials. We also show an altered expression sequence of Sox9 and Pax7, a border specifier gene. Our results thus provide a first clue to understand the molecular basis of the heterochronic development of CNC cells and jaw elements in marsupials.
MATERIALS AND METHODS
B6 mice were purchased from CLEA Japan, and midday of the vaginal plug was designated as embryonic day 0.5 (E0.5). Mouse embryos were staged according to Theiler’s stage (Theiler 1972). The breeding colony of gray short‐tailed opossum, M. domestica, is maintained at Nihon University School of Dentistry at Matsudo, under the approval of Nihon University Animal Care and Use Committee (No. AP09MD023, AP12MD015). M. domestica embryos were obtained from pregnant females, with mating video‐recorded and visually confirmed. M. domestica embryos were staged according to Mate et al. (1994), Vaglia and Smith (2003), and Smith (2006), which are adopted from McCrady stages (1938) established for Didelphis virginiana embryos. Fertilized Madagascar ground gecko (Paroedura pictus) eggs were obtained from a colony maintained at Kyoto Prefectural University of Medicine, and embryos were staged according to Noro et al. (2009). The Committee for Animal Experiment of Tohoku University Graduate School of Medicine approved the experimental procedures in this study.
In situ hybridization Whole‐mount and section in situ hybridizations were performed as described previously (Wakamatsu and Weston 1997). The DNA fragments of M. domestica Sox9, Snail1, and Snail2 corresponding to the exons were PCR‐amplified from genomic DNA prepared from M. domestica adult liver. The cDNA fragments of M. domestica BMP2, BMP4, and Pax7 were PCR‐amplified from oligo (dT) primed neonatal brain cDNA pool. These sequences were subcloned into pBluescriptII (Agilent Technologies, Santa Clara, CA, USA). The identities of M. domestica genes were confirmed by DNA sequencing. The sequences of primers were; Sox9F: GGAAGACACTGGTGATCTCC, Sox9R: CTACTCAGGGACTCTAGCTC, Snail1F: ATCCTCAGGCACAGCAGATG, Snail1R: CTAGTGTATGTCTAGTGTAC, Snail2F: CCCCATATCTCTATGAAAGC, Snail2R: AATGTGTCCTTGAAGCAACC, BMP2F: TTTTCGGGAGCAGGTGCAGG, BMP2R: ACCCACATCCCTCCACAACC, BMP4F: GGGACCAGAGAAAACCCTGC, BMP4R: CCTTCTACCACCATCTCCTG, Pax7F: TAAGCAGGCAGGAGCCAATC, Pax7R: CAACTTCATCCGACGCTGTG. cDNAs of avian Sox9 (Sakai et al. 2006), Sox10 (Cheng et al. 2000), Snail2 (Endo et al. 2002), mouse Snail1 (Sefton et al. 1998), and Sox10 (Sakai et al. 2005) for cRNA probes were described previously. Chicken Pax7 (Matsunaga et al. 2001) and mouse Sox9 cDNAs were kind gifts from Drs. Harukazu Nakamura and Ryohei Sekido, respectively. Gecko Sox9 cDNA was kindly provided by Drs. Miyuki Noro and Koji Tamura at Tohoku University (unpublished).
Experimental animals Chicken (Gallus gallus domesticus) eggs were obtained from a local farm. Embryos were staged according to Hamburger and Hamilton (HH stage, Hamburger and Hamilton 1951). Pregnant
Antibodies and immunostaining Anti‐Sox10 goat polyclonal antibody was commercially obtained (R&D systems). HNK1 (mouse IgM, Tucker et al.
Wakamatsu et al.
1988) was used as described previously (Wakamatsu et al. 2004). Fluorochrome‐conjugated secondary antibodies were purchased from Jackson Immuno Research. Immunological staining on cryo‐sections was performed as described previously (Wakamatsu et al. 1993, 1997). Sections treated with the antibody were also exposed to DAPI (Sigma‐ Aldrich, St. Louis, MO, USA) to visualize nuclei.
RESULTS AND DISCUSSION
Embryonic stages chosen for this study M. domestica embryos used in this study were staged according to previously published papers (Mate et al. 1994; Vaglia and Smith 2003; Smith 2006). Stage 18–19 embryos appeared as flat neurulae with a small area of anterior neural plate with a long primitive streak, morphologically corresponding to stage 3–4 chicken embryos. Stage 20 M. domestica embryos had anteriorly elongated neural plate with a sign of head neural fold elevation with no somites (Fig. 1a). From stage 21 to 22, the elevation of the head neural folds became apparent, and the anterior neural plate expanded laterally. In stage 22 embryos, at the lateral edges of the head neural folds, preotic and otic sulci were apparent, and a few somite pairs were formed. Embryos of this stage also showed apparent anterior–posterior elongation, and the lateral expansion of the anterior neural plate likely corresponded to the presumptive CNC region. At stage 23, a mass of delaminated CNC cells was observed (Fig. 1b, see also Smith 2006). At this stage, in addition to preotic and otic sulci, an additional notch was observed more anteriorly (designated as “anterior neural sulcus” in this study). From stage 24 to 26, the maxillo‐ mandibular and hyoid arches gradually became evident (Fig. 1c). At this stage, the elevation of neural folds was also apparent. By stage 28–29, cranial anlagen, such as fronto‐nasal mass, maxillary process, mandibular arch, and hyoid arch, as well as
Sox9 in marsupial neural crest
199
the olfactory placode, have enlarged (Fig. 1d). At these stages, the neural folds are dorsally fused along the antero‐posterior neural axis. Large fore limb buds and small hind limb buds were also observed.
Expression patterns of Snail1 in M. domestica Snail family transcription factors have been known to promote EMT of a broad range of cell types including NC. In chicken and quail embryos, Snail2 is expressed both in the CNC and trunk NC, and the expression begins prior to EMT, while Snail1 is not expressed in the NC (Sefton et al. 1998; Sakai et al. 2006). Snail2 has been known to regulate cell adhesion (Cheung and Briscoe 2003; Cheung et al. 2005; Taneyhill et al. 2007), and both blocking Snail2 activity by antisense RNA or by a dominant‐negative Snail2 inhibit EMT of CNC in vivo and in vitro, respectively (Nieto et al. 1994; Sakai et al. 2006). In mouse embryos, both Snail1 and Snail2 are expressed in migrating NC, but only Snail1 appears to be expressed prior to the EMT (Sefton et al. 1998; Sakai et al. 2005). Thus, we have cloned M. domestica homologs of both Snail1 and Snail2, and examined their expression. Snail2 expression was not observed in the CNC around the time of delamination (data not shown), hence we focused on Snail1 expression. At embryonic stage 20, Snail1 expression was strongly observed in the primitive streak (Fig. 2a), as well as in the mesodermal cells surrounding the neural plate (Fig. 2a), but no Snail1 expression was detected in the ectodermal components including the prospective CNC area. At stage 23, a bilateral expression domain of Snail1 was observed, likely corresponding to the CNC cells (Fig. 2b). Sections revealed Snail1 expression both in the CNC cells and in the mesodermal components (Fig. 2, c and d). At stage 26, when the cranial anlagen became evident, Snail1 expression was clearly observed in the CNC cells of the maxillo‐ mandibular and hyoid arches (Fig. 2, e and f), and this expression
Fig. 1. Morphologies of gray short‐tailed opossum (Monodelphis domestica) embryos examined in this study. (a) Stage 20 embryo, showing anterior neural plate (np), node (nd), and primitive streak (ps). (b) A dorsal view of stage 23 embryo. Seven pairs of somite (so) can be observed. While anterior neural plate is wide open, the CNC cells have already started generating cranial anlagen (arrowheads). Anterior neural sulcus (ans), preotic sulcus (pos), and otic sulcus (os) can be morphologically identified. (c) A dorsal view of stage 26 embryo. The cranial and trunk neural folds have not been fused yet. The maxillo‐mandibular (mx‐mb) and hyoid (hy) arches are apparent. (d) A lateral view of stage 29 embryo. The fronto‐nasal (fn), maxillary (mx), mandibular (md), hyoid (hy) processes are evident. Huge fore limb bud (fl) and small hind limb bud (hl) are also observed. Scale bars: a, 400 mm; b and c, 200 mm; d, 1 mm.
200
EVOLUTION & DEVELOPMENT
Vol. 16, No. 4, July–August 2014
Early, broad, and persistent expression patterns of Sox9 in M. domestica
Fig. 2. Expression of Snail1 mRNA in M. domestica embryos. (a) Dorsal view of a stage 20 embryo. While strong expression of Snail1 in the primitive streak (ps) and a weaker expression in the lateral mesoderm are observed, little expression is detectable in the CNC domain (arrow). (b) A dorsal view of stage 23 embryo. The expression of Snail1 in the CNC cells is clearly observed (arrowhead). The expression in the somites (so) is also detected. The planes of sections shown in (c) and (d) are indicated. (c) A transverse section of the embryo shown in (b). Snail1 expression is evident in the CNC cells (arrowheads) and paraxial mesoderm (mes), but not in the neural plate (np). (d) A transverse section of the embryo shown in (b). Snail1 expression is detected in the somite (so). (e) A dorsal view of stage 26 embryo. The cranial processes such as maxillo‐mandibular (mx‐md) and hyoid (hy) arches strongly express Snail1. In addition to the expression in the somites and primitive streak, the lateral plate mesoderm corresponding to the future fore limb bud (fl) also expresses Snail1. (f) A lateral view of embryo shown in (e). Snail1 expression in maxillo‐mandibular (mx‐ md) and hyoid (hy) arches is evident. The anterior neural sulcus (see also Fig. 1b) is indicated by an arrowhead. (g) A section of stage 29 embryo at the hindbrain level. Snail1 is expressed homogenously in the CNC cell‐derived ectomesenchyme of the mandibular arch (md). Scale bars: a, 400 mm; b and e, 200 mm.
continued at least until stage 29 (Fig. 2g). We also noted that mesenchymal cells rostral to the anterior neural sulcus were negative for Snail1 expression (Fig. 2f). Because these cells expressed Sox9 (see below), and because Snail1 is expressed in the maxillio‐mandibular CNC cells but not in the fronto‐nasal CNC cells of mouse embryos (Sefton et al. 1998), these cells likely corresponded to the fronto‐nasal population of CNC cells of M. domestica embryos. Snail1 expression in the fore limb field and the paraxial mesoderm was also similar to that in mouse embryos (Sefton et al. 1998. See also Keyte and Smith 2010 for M. domestica limb development), thus the expression pattern of Snail1 seemed to be largely conserved in M. domestica and mouse.
Group E Sox transcription factor genes, such as Sox9 and Sox10, are important for the promotion of EMT and differentiation of NC (See reviews mentioned above). In particular, Sox9 has been shown to be crucial for the reduction of cell–cell adhesion during EMT of NC, the growth of cranial processes, and the differentiation of cartilage in mouse and avian systems (Lefebvre et al. 1997; Mori‐Akiyama et al. 2003; Akiyama et al. 2004; Suzuki et al. 2006; Sakai et al. 2006). We thus examined the expression of Sox9 in M. domestica embryos. At stage 18, no bilateral expression of Sox9 was observed (Fig. 3a). The onset of Sox9 expression in the prospective CNC was, therefore, around stage 19 (Fig. 3b), prior to neural fold elevation, and well prior to segmentation of the first somite. Observation on whole‐mount and sections revealed that, at stage 20, the Sox9‐positive cells still remained as a part of the ectodermal epithelium prior to EMT (Fig. 3, c and d). At stage 23, many Sox9‐positive CNC cells have delaminated, and consisted of virtually all the mesenchyme cells of the fronto‐ nasal and maxillo‐mandibular processes (Fig. 3, e and f). At stage 25, Sox9‐positive cells still occupied the mesenchyme of the cranial processes including the hyoid arches (Fig. 3, g and h). Unlike Snail1, Sox9‐positive CNC cells were clearly found anteriorly to the anterior neural sulcus, consistent with these cells being the fronto‐nasal CNC cells (Fig. 3h). At stage 28, Sox9 expression was still observed in the fronto‐nasal, maxillary, mandibular processes, as well as hyoid arches (Fig. 3i). At stage 29, Sox9 expression was largely lost in the fronto‐ nasal, mandibular and hyoid, but remained in the maxillary processes (Fig. 3, j, k, l). From stage 20–29, we noted the expression of Sox9 in more posterior NC including trunk NC (Fig. 3, c, e, g, j). We also examined Sox10 protein expression in M. domestica by immunostaining (Fig. 4). At stage 20, The CNC cells forming the cranial anlagen strongly expressed Sox10 (Fig. 4, a–a00 ). At stage 26, the CNC‐derived ectomesenchymal cells in the cranial processes such as the mandibular arch expressed Sox10 protein, but the expression level was lower compared to the more dorsally positioned CNC cells (Fig. 4, b–b00 ). Based on their location, these dorsal CNC cells were likely to be progenitors of the cranial ganglia (see below). At stage 28, Sox10 expression was barely detectable in the CNC‐derived ectomesenhcyme of the cranial anlagen, while high Sox10 expression was clearly observed in the cranial ganglia (Fig. 4, c–c00 ). These observations indicated that CNC cells co‐expressed Sox9 and Sox10 for a long period, but Sox10 expression diminished earlier than Sox9 in the cranial ectomesenchyme. The persistent expression of Sox9 in CNC cells and co‐ expression of Sox9 and Sox10 in the CNC‐derived ectomesenchyme of the cranial processes in M. domestica seemed to be different from those in avian embryos. We have previously
Wakamatsu et al.
Sox9 in marsupial neural crest
201
Fig. 3. Expression of Sox9 mRNA in NC cells of M. domestica embryos. (a–c) A series of M. domestica embryos prior to the CNC cell delamination. At stage 18, Sox9 expression is weakly and broadly detected in the neural plate (a). At stage 19, Sox9 expression is upregulated bilaterally in the presumptive CNC domain (b, arrowheads). At stage 20, Sox9 expression in the CNC is evident (c, arrowhead). Premigratory NC cells in the more posterior region are also recognized by Sox9 expression (arrows). (d) A transverse section of the embryo shown in (c) at the indicated level. Sox9 expression in the premigratory CNC (arrowheads) is observed bilaterally to the neural plate (np). (e) A dorsal view of stage 23 embryo, showing Sox9 expression both in the CNC (arrowhead) and the trunk NC (arrows). Sox9 expression is also observed in the somites (so). The preotic sulcus (pos) is indicated. The axial level corresponding to sections shown in (f) is also indicated. (f) A transverse section of a stage 23 embryo similar to the embryo in (e). Sox9 mRNA is expressed in the CNC cells forming the maxillo‐mandibular anlagen (arrowheads), as well as in the mesodermal cells (mes). (g and h) Dorsal (g) and lateral (h) views of a stage 25 embryo. Sox9 expression is evident in the CNC cells of fronto‐nasal (fn), maxillo‐mandibular (mx‐md), and hyoid (hy) processes. Sox9 expression is also detected in the otic vesicle (ov), somites, and trunk NC (arrows in g). Some premigratory CNC cells remain at the lateral edge of the head neural fold (arrows in h). Anterior neural sulcus is indicated (arrowhead in h). (i) Side view of the cranial region of a stage 28 embryo. Sox9 is expressed in the fronto‐nasal (fn), maxillary (mx), mandibular (md), and hyoid (hy) processes. (j) Side view of a stage 29 embryo. Trunk NC cells express Sox9 (arrows). fl: fore limb bud. (k) High magnification view of the head of the embryo, shown in (j). The expression of Sox9 is significantly reduced in the cranial anlagen except for the maxillary process (mx). (l) Section of stage 29 embryo at the level of the hindbrain, showing Sox9 expression. Note the lack of Sox9 mRNA in the distal‐medial region of the mandibular arch (md). Scale bars: a–c, e, g, 200 mm; i–k, 1 mm.
shown that, in chicken and quail embryos, Sox9 expression in the head neural folds was first observed at stage 7 (1 somite), and the expression was up‐regulated in the premigratory CNC cells by stage 8 (4 somites), but Sox9 expression was quickly down‐ regulated at the early phase of CNC migration (see Suzuki et al. 2006; Sakai et al. 2006). This down‐regulation coincided with Sox10 expression, probably because Sox10 could repress Sox9, while Sox9 could induce Sox10 (Suzuki et al. 2006). Because detailed observation of Sox9/Sox10 expression has not been done in the CNC of eutherian mammals, we also performed whole‐mount in situs of mouse embryos from E8.0 to E8.75 (Theiler stage 11d‐13, Fig. 5) for further comparisons. Neither Sox9 nor Sox10 was expressed in the neural folds of the pre‐ somitegenesis stage embryos at E8.0 (Theiler stage 11d, Fig. 5, a and f), and Sox9 expression was first turned on in the premigratory CNC of the cranial neural folds around E8.0‐
8.25 (Theiler stage 12a, Fig. 5b). Thus, the onset of Sox9 expression in chicken and mouse was relatively later than in M. domestica if somitegenesis was used as a standard of developmental stage. Sox9 expression was evident in the mouse CNC cells at the early phase of migration (Theiler stage 12a, Fig. 5c), but diminished during migration (Theiler stage 12b, Fig. 5d), and was turned off in the CNC cells entering the cranial processes at 8.75 dpc (Theiler stage 13, Fig. 5e). In mouse embryos at similar stages, Sox10 was weakly detected in the premigratory CNC cells at E8.25 (Theiler stage 12a, Fig. 5g), was upregulated in the migrating CNC cells at E8.25–8.5 (Theiler stage 12a–12b, Fig. 5, h and i), and persisted in the CNC cells forming the cranial processes at E8.5 (Theiler stage 13, Fig. 5j). Thus, the transition of the group E Sox genes from Sox9 to Sox10 in the mouse CNC cells resembled to that in chicken, but not to that in M. domestica.
202
EVOLUTION & DEVELOPMENT
Vol. 16, No. 4, July–August 2014
Fig. 4. Expression of Sox10 protein in CNC cells of M. domestica embryos. Transverse sections of stage 23 (a–a00 ), stage 26 (b–b00 ), and stage 28 (c–c00 ) embryos at the level of the mandibular arch are shown. DAPI staining shows the nuclei. (a–a00 ) At stage 23, the CNC cells strongly express Sox10 protein (arrowheads), but not in other tissues, such as neural plate (np) and mesoderm (mes). (b–b00 ) At stage 26, Sox10 expression is detected in the CNC‐derived ectomesenchyme (arrowheads) of the mandibular arch (md). Note that the staining intensity in the ectomesenchyme is lower than that in the late migrating CNC cells remaining dorsally ( ). (c–c00 ) At stage 28, while sensory ganglia strongly express Sox10 protein ( ), the expression is barely detectable in the mandibular ectomesenchyme (arrowheads).
It is possible that the persistent expression of Sox9 in M. domestica might be a unique feature acquired in the marsupial lineage. Alternatively, the persistent expression of Sox9 might be an ancestral feature for mammals, and that the similar sequential expression of Sox9/Sox10 in chicken and mouse might have been established independently, as a result of an evolutionary convergence. To explore this further, we examined the expression of Sox9 in gecko embryos, as an example of a reptile (Fig. 6). At 0.5 dpo (days postoviposition), Sox9 expression was evident along the dorsal midline of the midbrain as the left and right neural folds were fused (Fig. 6, a and b). The CNC identity of these Sox9‐positive cells was confirmed by HNK1 staining on sections (Fig. 6, b–g). Thus, the late onset of the CNC cell migration at the time of cranial neural fold closure appears similar in gecko and chicken embryos, as well as in crocodile embryos as previously reported (Kundrat 2009), but not in mouse or M. domestica embryos, in which the CNC cells started migrating well before the fusion of the head neural folds (Figs. 3, e, f, and 5c). A stream of Sox9 expression from the caudal end of the dorsal midline of the midbrain was also observed in gecko embryos (Fig. 6, a, e, f, h, i), suggesting that,
Fig. 5. Expression of Sox9 and Sox10 mRNA in CNC cells of mouse embryos. The mRNA expression of Sox9 (a–e) and Sox10 (f–j) are shown from E8 to E8.75 (Theiler stage 11d‐13). An intense Sox9 expression is detected earlier than Sox10 expression in the premigratory CNC along the lateral edge of the cranial neural folds (arrowheads; compare (a, b) and (f, g)). Sox9 expression in the CNC cells is downregulated during migration (d), while Sox10 expression persists (i). At the oldest stage shown in this figure, Sox9 expression is largely absent in the cranial processes (e), although Sox9 expression is detectable in the otic placode (op) and the trunk NC cells (arrows in e). At similar stage, Sox10 expression is observed in all the cranial processes, such as fronto‐nasal (fn), mandibular (md), and hyoid (hy), as well as in the trunk NC cells (arrows in j). Scale bars: a and f, 100 mm; b–e and g–j, 200 mm.
Fig. 6. Expression of Sox9 mRNA in CNC cells of gecko embryos. (a) Side view of 0.5 days postoviposition (dpo) gecko embryo, showing Sox9 mRNA expression in the CNC cells (arrowheads) and otic placode (op). A large arrowhead indicates CNC cells on top of the fore‐midbrain. Small arrowheads indicate posterior migratory stream of the midbrain CNC cells. (b–d) A section of a Sox9 in situ hybridized embryo shown in (a), counter‐stained with HNK1 antibody and DAPI. The plane of section is indicated in (a). HNK1‐ positive CNC cells reside on top of the midbrain (large arrowhead). (e–g) A section of (a), counter‐stained with HNK1 antibody and DAPI. The plane of section is indicated in (a). Most CNC cells in the posterior migratory stream are Sox9 and HNK1‐double positive (small arrowhead), although the Sox9‐positive CNC cells residing in the most distal part of the stream appears to be HNK1‐negative. (h) A side view of the head of 0.5 dpo embryo (slightly older than the embryo shown in (a). Only a few Sox9‐positive CNC cells remain on the top of midbrain (an arrow), and many Sox9‐positive CNC cells occur in the posterior migratory stream (arrowheads). (i) Side view of the head of 1.5 dpo embryo. While Sox9 expression remains in the posterior migratory stream (arrowheads), Sox9 expression is absent in the maxillo‐mandibular arches (mx‐md). Scale bars: a, 2 mm; b and c, 1 mm.
Wakamatsu et al.
Fig. 7. Expression of BMP2/4 mRNA in chicken, mouse, and M. domestica embryos at stages prior to CNC cell migration. (a–c) BMP4 mRNA expression in chicken embryos at stage 5 (a), stage 6 (b), and stage 8.5 (c). At stage 5, BMP4 expression is detected in the non‐neural ectoderm surrounding the neural plate (arrowheads in a). At later stages, BMP4 expression in the non‐neural ectoderm is downregulated, while strong expression along the neural folds is observed (arrows in b, c). BMP4 is also expressed in the primitive streak (ps) of chicken embryos (a–c). (d) BMP2 expression in the non‐neural ectoderm surrounding the neural plate of E8.0 (Theiler stage 11d) mouse embryo (dorsal view, arrowheads). (e–h) BMP2 mRNA expression in M. domestica embryos at stage 19 (e), stage 20 (f), stage 21 (g), and stage 22 (h). BMP2 expression is observed in the non‐neural ectoderm surrounding the neural plate (arrowheads) and in the cranial neural folds of older embryos (arrows). Scale bars: a–d, 100 mm; e–h, 200 mm.
unlike the CNC cells of chicken, mouse, or M. domestica, in which the midbrain CNC cells migrate broadly toward the position of the future maxillo‐mandibular anlage, the gecko midbrain CNC cells appeared to have a more restricted migratory pathway. This posterior migratory stream of the midbrain CNC has been previously identified in crocodile embryos (Kundrat 2009). It seemed likely that the posterior migratory pathway of the CNC cells found in lepidosaurian gecko and archosaurian crocodile represents the ancestral state of reptilia, and that the broad migratory stream of midbrain CNC cells in mammals and archosaurian chicken would be the result of the evolutionary convergence. Nevertheless, in 0.5 and 1.5 dpo gecko embryos, Sox9 expression was not found more ventrally in the cranial anlagen (Fig. 6, h and i), suggesting that, in the gecko CNC cells, the Sox9 expression was transient, similar to that in chicken and mouse embryos, but not to that in M. domestica. Thus, the persistent expression of Sox9 in the ectomesenchymal
Sox9 in marsupial neural crest
203
Fig. 8. Pax7 expression in chicken (a–e) and M. domestica (f–i) embryos. (a–e) From stage 5 to 7, Pax7 expression is observed along the neural border of chicken embryos (arrows in a–d). At stage 7, the anterior part of the Pax7‐positve border domain slightly expands (d). At stage 8.5, Pax7 is strongly expressed in CNC cells at the onset of their migration (arrows in e). The midbrain‐hindbrain boundary (mhb) is also indicated. (f) At the earliest stage of M. domestica embryo examined (stage 18), Pax7 expression is only faintly observed in the neural plate. (g) At stage 20, Pax7 expression is detectable along the border region (arrows). (h) At stage 21.5, while Pax7 expression along the border mostly remains as a lateral thin line (arrows), in the premigratory CNC region, the Pax7 expression domain expands medially (arrowheads). Preotic sulcus (pos) is indicated. (i) A high magnification view of the anterior part of the embryo in (h) showing lateral high‐medial low gradient of Pax7 expression in the premigratory CNC domain (arrowheads). Arrows indicate the Pax7 expression along the lateral edge of the neural fold. Scale bars: a–e, 100 mm; f–h, 200 mm.
CNC cells of M. domestica embryos seemed to be a peculiarity of development of the cranial anlagen in Marsupials rather than an ancestral feature.
Expression patterns of BMPs in M. domestica BMP signaling is essential for CNC induction in various amniotes (Kanzler et al. 2000; Endo et al. 2002; see also reviews listed above). In chicken embryos, BMP4 was initially expressed in the non‐neural ectoderm at stage 5, well prior to the onset of Sox9 expression in the premigratory CNC (Fig. 7a), and as the non‐neural ectoderm expression gradually diminished, the cranial neural folds expressed BMP4 at stage 7 and 8.5 (Fig. 7, b and c), coincided with the elevation of Sox9 expression in the premigratory CNC cells (see above). In mouse embryos, BMP2 was expressed in the non‐neural ectoderm at E8.0 (Theiler stage 11d, Fig. 7d; see also Kanzler et al. 2000), prior to the onset of Sox9 expression in the premigratory CNC. Thus, we have cloned the M. domestica homologs of BMP2 and BMP4, and
204
EVOLUTION & DEVELOPMENT
Vol. 16, No. 4, July–August 2014
Table 1. Comparisons of Pax7 expression, Sox9 expression, and the migratory behavior of CNC cells in mouse, opossum, gecko, and chicken embryos
Pax7 expression
The onset of Sox9 expression
The diminishing point of Sox9 expression
The onset of CNC cell migration
Migratory behavior of midbrain CNC cells
Mouse
Neural folds (Basch et al. 2006)
Later than the first somite stage, at the middle phase of neural fold elevation (this study)
Early: during migration (this study)
Middle phase of neural fold elevation (this study)
Anterior‐posteriorly unrestricted (Osumi‐ Yamashita et al. 1997)
Opossum
Initially surrounding the anterior neural plate, then bilateral bands (this study)
Late: in the cranial processes (this study)
n.d.
Chicken (and quail)
Bilateral bands (Basch et al. 2006, this study)
At the onset of somitegenesis, at the middle phase of neural fold elevation (Sakai et al. 2006)
Early: during migration (Suzuki et al. 2006)
Beginning of neural fold elevation (This study. See also Vaglia and Smith 2003) Neural fold closure (This study. See Kundrat 2009, for crocodile) Neural fold closure (This study. See also Kundrat 2009, for ostrich)
Anterior‐posteriorly unrestricted (Vaglia and Smith 2003, this study)
Gecko
Before the first somite stage, at the beginning of neural fold elevation (this study) n.d.
Early: during migration (this study)
Posterior migratory stream (This study. See Kundrat 2009, for crocodile) Anterior‐posteriorly unrestricted (see Le Douarin and Kalcheim 1999)
n.d., Not determined.
examined their expression patterns (Fig. 7, e–h). In M. domestica embryos, BMP4 expression was not detected in the stage 20–23 embryos, although at the trunk level of stage 29 embryos, dorsal neural tube expression was observed, similar to that in mouse and chicken embryos (data not shown). In contrast, M. domestica BMP2 was detected in the non‐neural ectoderm at stage 19 (Fig. 7e). In older embryos such as stage 20, 21, and 22, while BMP2 expression in the non‐neural ectoderm gradually regressed peripherally (Fig. 7, f–h), BMP2 expression along the lateral edge of the neural folds became evident (Fig. 7, f–h).
Because this expression pattern of BMP2 in M. domestica embryos appeared to be similar to BMP4 in chicken embryos, BMP2, but not BMP4, might be involved in CNC formation of M. domestica.
Expression patterns of Pax7 in M. domestica Previous studies in chicken, frog, and zebrafish embryos suggested that, prior to the induction of “neural crest specifier” genes such as Snail1/Snail2 and Sox9, the non‐neural/neural
Fig. 9. Diagrammatic comparisons of morphological events, developmental status of the CNC, and Sox9/Sox10 expression in the CNC among chicken, mouse and opossum. Developmental stages are also indicated. Note the early onset of the Sox9 expression, the cranial anlagen protrusion, and the late fusion of neural fold in the opossum. (Note the onset of the Sox10 expression in the opossum is not determined.)
Wakamatsu et al.
border has been established by a combination of inductive signals (Milet and Monsoro‐Burq 2012, for a review). In chicken embryos, Pax7 expression was a good indicator of border specification, and Pax7 function was required for the expression of NC specifier genes, including Sox9 (Basch et al. 2006). Thus, we compared Pax7 expression in chicken (Fig. 8, a–e) and M. domestica embryos (Fig. 8, f–i). At stage 18 of M. domestica embryo, prior to Sox9 expression, Pax7 expression was not detected at the border along the newly formed neural plate (Fig. 8f). At stage 20, Pax7 expression was observed in the neural border (Fig. 8g) in a very similar manner to that in stage 4– 7 chicken embryos (Fig. 8, a–c). However, this expression pattern was different from the broad, bilateral expression domain of Sox9 in the prospective CNC at similar stages of M. domestica embryos (see Fig. 3, b and c). In addition, the onset of Pax7 expression in the border was no earlier than Sox9 expression in the premigratory CNC (compare Figs. 3, a–c and 8, f and g). Thus, the inductive sequence of the non‐neural/neural border specification demarcated by Pax7 expression and the subsequent CNC induction indicated by the bilateral expression of Sox9 found in chicken is difficult to apply to M. domestica embryos. At older stages of M. domestica embryos such as stage 21.5, a lateral expansion of the Pax7 domain was observed in the area corresponding to the premigratory CNC expressing Sox9 (Fig. 8, h and i; see also Fig. 3). Interestingly, this Pax7 expression showed a “lateral high‐medial low” gradient (Fig. 8i). We thus speculate that Pax7 expression in CNC would be under the influence of medio‐lateral graded signals, such as BMP2 expressed at the lateral edge of the neural folds. Nonetheless, the distinct expression sequence of Pax7 and Sox9 in M. domestica suggested the altered epistatic relationship between these genes, and its contribution to heterochronic formation of the CNC.
CONCLUSION In this study, we examined gene expression patterns in the neural plate and CNC cells of M. domestica embryos, and correlated them with the early developmental events and the process of the CNC development. Our results are summarized in Table 1 and Fig. 9. Sox9 expression in M. domestica CNC cells is apparently modified, compared to other amniotes. The early and broad expression of Sox9 in the premigratory CNC domain clearly corresponds to, and likely contributes to, the early generation of many CNC cells shown in the previous histological studies (Smith 2001; Vaglia and Smith 2003; Smith 2006). Although the role of the persistent expression of Sox9 in the CNC cell‐derived ectomesenchyme of cranial processes remains unclear, it may also contribute to the cartilage differentiation of the ectomesenchyme. This study also indicated that, because gecko (this study) and crocodile (Kundrat 2009) CNC cells at the midbrain level have a conserved posterior migratory stream, the lack of this
Sox9 in marsupial neural crest
205
migratory stream found in the avian and mammalian embryos might have evolved independently. Acknowledgments We thank Drs. Koji Tamura and Don Newgreen for comments on the manuscript. We thank Drs. Harukazu Nakamura, Ryohei Sekido, Koji Tamura, and Miyuki Noro for plasmids. We also thank Noriko Uemura for opossum care. This work was supported in part by a grant to Y.W. from JSPS (KAKENNHI Grant Number 24570228), and to K.S. (KAKENNHI Grant Number 20592154).
REFERENCES Akiyama, H., et al. 2004. Essential role of Sox9 in the pathway that controls formation of cardiac valves and septa. Proc. Natl. Acad. Sci. 101: 6502– 6507. Basch, M. L., Bronner‐Fraser, M., and Garcia‐Castro, M. I. 2006. Specification of the neural crest occurs during gastrulation and requires Pax7. Nature 441: 218–222. Betancur, P., Bronner‐Fraser, M., and Sauka‐Spengler, T. 2010. Assembling neural crest regulatory circuits into a gene regulatory network. Annu. Rev. Cell. Dev. Biol. 26: 581–603. Cheng, U., Cheung, M., Abu‐Elmagd, M. M., Orme, A., and Scotting, P. J. 2000. Chick sox10, a transcription factor expressed in both early neural crest cells and central nervous system. Brain Res. Dev. Brain Res. 121: 233–241. Cheung, A. F. P., et al. 2009. The subventricular zone is the developmental milestone of a 6‐layered neocortex: comparisons in metatherian and eutherians mammals. Cereb. Cortex 20: 1071–1081. Cheung, M., and Briscoe, J. 2003. Neural crest development is regulated by the transcription factor Sox9. Development 130: 5681–5693. Cheung, M., Chaboissier, M. C., Mynett, A., Hirst, E., Schedl, A., and Briscoe, J. 2005. The transcriptional control of trunk neural crest induction, survival, and delamination. Dev. Cell. 8: 179–192. Debora, C. K., and Sears, K. E. 2010. The divergent development of the apical ectodermal ridge in the marsupial Monodelphis domestica. Anat. Rec. 290: 1325–1332. Endo, Y., Osumi, N., and Wakamatsu, Y. 2002. Bimodal functions of Notch‐ mediated signaling are involved in neural crest formation during avian ectoderm development. Development 129: 863–873. García‐Castro, M., Marcelle, C., and Bronner‐Fraser, M. 2002. Ectodermal Wnt function as a neural crest inducer. Science 297: 848–851. Hall, B. K. 2000. The neural crest as a fourth germ layer and vertebrates as quadroblastic not triploblastic. Evol. Dev. 2: 3–5. Hamburger, V., and Hamilton, H. L. 1951. A series of normal stages in the development of the chick embryo. J. Morph. 88: 49–92. Kanzler, B., Foreman, R. K., Labosky, P. A., and Mallo, M. 2000. BMP signaling is essential for development of skeletogenic and neurogenic cranial neural crest. Development 127: 1095–1104. Keyte, A. L., and Smith, K. K. 2010. Developmental origins of precocial forelimbs in marsupial neonates. Development 137: 4283–4294. Krispin, S., Nitzan, E., Kassam, Y., and Kalcheim, C. 2010. Evidence for a dynamic spatiotemporal fate map and early fate restrictions of migratory avian neural crest. Development 137: 585–595. Kundrat, M. 2009. Heterochronic shift between early organogenesis and migration of cephalic neural crest cells in two divergent evolutionary phenotypes of archosaurs: crocodile and ostrich. Evol. Dev. 11: 535–546. Le Douarin, N. M., and Kalcheim, C. 1999. The Neural Crest. 2nd Ed. Cambridge University Press, Cambridge. Lefebvre, V., Huang, W., Harley, V. R., Goodfellow, P. N., and de Cromburgghe, B. 1997. SOX9 is a potent activator of the chondrocyte‐ specific enhancer of the proa1(II) collagen gene. Mol. Cell. Biol. 17: 2336–2346. Mate, K. E., Robinson, E. S., Vandeberg, J. L., and Pedersen, R. A. 1994. Timetable of in vivo embryonic development in the gray short‐tailed opossum (Monodelphis domestica). Mol. Rep. Dev. 39: 365–374.
206
EVOLUTION & DEVELOPMENT
Vol. 16, No. 4, July–August 2014
Matsunaga, E., Araki, I., and Nakamura, H. 2001. Role of Pax3/7 in the tectum regionalization. Development 128: 4069–4077. McCrady, E. 1938. The Embryology of the Opossum. Wistar Institute of Anatomy and Biology, Philadelphia. p. 233. McKeown, S. J., Wallace, A. S., and Anderson, R. B. 2013. Expression and function of cell adhesion molecules during neural crest migration. Dev. Biol. 373: 244–257. Milet, C., and Monsoro‐Burq, A. H. 2012. Neural crest induction at the neural plate border in vertebrates. Dev. Biol. 366: 22–33. Mori‐Akiyama, Y., Akiyama, H., Rowitch, D. H., and de Crombrugghe, B. 2003. Sox9 is required for determination of the chondrogenic cell lineage in the cranial neural crest. Proc. Natl. Acad. Sci. 100: 9360–9365. Moustakas, J. E., Smith, K. K., and Hlusko, L. 2011. Evolution and development of the mammalian dentition: insights from marsupial Monodelphis domestica. Dev. Dyn. 240: 232–239. Nieto, M. A., Sargent, M. G., Wilkinson, D. G., and Cooke, J. 1994. Control of cell behavior during vertebrate development by Slug, a zinc finger gene. Science 264: 835–839. Noro, M., Uejima, A., Abe, G., Manabe, M., and Tamura, K. 2009. Normal developmental stages of the Madagascar ground Gecko Paroedura pictus with special reference to limb morphogenesis. Dev. Dyn. 238: 100–109. Osumi‐Yamashita, N., Ninomiya, Y., and Eto, K. 1997. Mammalian craniofacial embryology in vitro. Int. J. Dev. Biol. 41: 187–194. Sakai, D., Tanaka, Y., Endo, Y., Osumi, N., Okamoto, H., and Wakamatsu, Y. 2005. Regulation of Slug transcription in embryonic ectoderm by beta‐ catenin‐Lef/Tcf and BMP‐Smad signaling. Dev. Growth Differ. 47: 471– 482. Sakai, D., and Wakamatsu, Y. 2005. Regulatory mechanisms for neural crest formation. Cells Tissues Organs 179: 24–35. Sakai, D., Suzuki, T., Osumi, N., and Wakamatsu, Y. 2006. Cooperative action of Sox9, Snail2, and PKA signaling in early neural crest development. Development 133: 1323–1333. Sefton, M., Sanchez, S., and Nieto, M. A. 1998. Conserved and divergent roles for members of Snail family of transcription factors in the chick and mouse embryo. Development 125: 3111–3121. Smith, K. K. 2001. Early development of the neural plate, neural crest and facial region of marsupials. J. Anat. 199: 121–131.
Smith, K. K. 2006. Craniofacial development in marsupial mammals: developmental origins of evolutionary change. Dev. Dyn. 235: 1181– 1193. Stuhlmiller, T. J., and García‐Castro, M. I. 2012. Current perspectives of the signaling pathways directing neural crest induction. Cell. Mol. Life Sci. 69: 3715–3737. Suzuki, T., Sakai, D., Osumi, N., Wada, H., and Wakamatsu, Y. 2006. Sox genes regulate type 2 collagen expression in avian neural crest cells. Dev. Growth Differ. 48: 477–486. Suzuki, T., Osumi, N., and Wakamatsu, Y. 2010. Stabilization of ATF4 protein is required for the regulation of epithelial–mesenchymal transition of the avian neural crest. Dev. Biol. 344: 658–668. Taneyhill, L. A., Coles, E. G., and Bronner‐Fraser, M. 2007. Snail2 directly represses cadherin6B during epithelial‐to‐mesenchymal transitions of the neural crest. Development 134: 1481–1490. Theveneau, E., and Mayor, R. 2012. Neural crest delamination and migration: from epithelium‐mesenchyme transition to collective cell migration. Dev. Biol. 366: 34–54. Tucker, G. C., Delarue, M., Zada, S., Boucaut, J. C., and Thiery, J. P. 1988. Expression of the HNK‐1/NC‐1 epitope in early vertebrate neurogenesis. Cell Tissue Res. 251: 457–465. Theiler, K. 1972. The House Mouse. Springer‐Verlag, New York. Vaglia, J. L., and Smith, K. K. 2003. Early differentiation and migration of cranial neural crest in the opossum, Monodelphis domestica. Evol. Dev. 5: 121–135. Wakamatsu, Y., Watanabe, Y., Shimono, A., and Kondoh, H. 1993. Transition of localization of the N‐myc protein from nucleus to cytoplasm in differentiating neurons. Neuron 10: 1–9. Wakamatsu, Y., and Weston, J. A. 1997. Sequential expression and role of Hu RNA‐binding proteins during neurogenesis. Development 124: 3449– 3460. Wakamatsu, Y., Watanabe, Y., Nakamura, H., and Kondoh, H. 1997. Regulation of the neural crest cell fate by N‐myc: promotion of ventral migration and neuronal differentiation. Development 124: 1953–1962. Wakamatsu, Y., Endo, Y., Osumi, N., and Weston, J. A. 2004. Multiple roles of SOX2, a HMG‐box transcription factor in avian development. Dev. Dyn. 229: 74–86.
16:4, 197–206 (2014)
DOI: 10.1111/ede.12083
Comparative gene expression analyses reveal heterochrony for Sox9 expression in the cranial neural crest during marsupial development Yoshio Wakamatsu,a Tadashi Nomura,b Noriko Osumi,a and Kunihiro Suzukic a
Department of Developmental Neuroscience, United Centers for Advanced Research and Translational Medicine (ART), Tohoku University Graduate School of Medicine, Sendai, Miyagi 980‐8575, Japan b Department of Biology, Kyoto Prefectural University of Medicine, Kyoto 603‐8334, Japan c Department of Biology, Nihon University School of Dentistry at Matsudo, Chiba 271‐8587, Japan *Author for correspondence (e‐mail: [email protected])
SUMMARY Compared to placental mammals, marsupials have short gestation period, and their neonates are relatively immature. Despite these features, marsupial neonates must travel from the birth canal to the teat, suckle and digest milk to complete development. Thus, certain organs and tissues of marsupial neonates, such as forelimbs to crawl and jaw elements to suckle, must develop early. Previous reports showed that cranial neural crest (CNC) cells, as the source of ectomesenchyme of jaw elements, are generated significantly early in gray short‐tailed opossum (Monodelphis domestica) compared to other amniote models, such as mouse. In this study, we examined the expression of genes known to be important for neural crest formation, such as BMP2/BMP4 (neural crest inducer), Pax7 (neural border specifier), Snail1 and Sox9/Sox10 (neural crest specifier) in Monodelphis
domestica, and compared the expression patterns with those in mouse, chicken, and gecko embryos. Among those genes, the expression of Sox9 was turned on early and broadly in the premigratory CNC cells, and persisted in the ectomesenchyme of the cranial anlagen in opossum embryos. In contrast, Sox9 expression diminished in the CNC cells of other animals at the early phase of migration. Comparison of the onset of Pax7 and Sox9 expression revealed that Sox9 expression in the prospective CNC was earlier and broader than Pax7 expression in opossum, suggesting that the sequence of border specification and neural crest specification is altered. This study provides the first clue for understanding the molecular basis for the heterochronic development of the CNC cells and jaw elements in marsupials.
INTRODUCTION
their mating has to be visually confirmed), and therefore has been used in a variety of evolutionary developmental studies including the heterochronic development of limbs (Keyte and Smith 2010; Debora and Sears 2010), the brain (Cheung et al. 2009), and the tooth (Moustakas et al. 2011). Completion of genome sequencing will also facilitate the use of M. domestica. In previous studies, Smith and colleagues have shown that cranial neural crest (CNC) cells of M. domestica embryos delaminate, and start forming cranial processes, such as mandibular, maxillary, and hyoid arches, relatively earlier compared to those of eutherian embryos (Vaglia and Smith 2003; see also Smith, 2001 and 2006 for reviews). It is not known, however, how this heterochronic development of the CNC is molecularly and genetically regulated. Neural crest (NC), which was referred to as “the fourth germ layer” specific in vertebrates (Hall 2000, and references therein), gives rise to a wide variety of tissues and cells types, including neurons and glial cells of the peripheral nervous system, pigment cells in the skin, and skeletal and connective tissues in the head (Le Douarin and Kalcheim 1999). Among those, the ectomesenchymal subpopulations specific to the CNC contribute
Marsupial (metatherian) mammals, such as kangaroos and koalas, have significantly short gestation periods compared to placental (eutherian) mammals, and their neonates are, thus, relatively altricial at birth (see Smith 2001 and references therein). Despite their immature state of developmental stage, marsupial neonates must travel form the birth canal to the teat, suckle, and digest milk to complete the rest of their development. Therefore, certain tissues and organs required for such behavior have to develop relatively early during their short embryonic period. Among such organs, craniofacial structures, such as jaws and throat, are important for suckling the milk and breathing, and previous morphological and histological studies by Smith and her colleagues have revealed the remarkably early, “heterochronic” development of the cranial anlagen in the embryo of the gray short‐tailed opossum, Monodelphis domestica (M. domestica, Vaglia and Smith 2003; see also Smith 2001, 2006 for reviews). M. domestica is favored as an experimental laboratory marsupial, because of its relatively small size (between mouse and rat), ease of cultivation in captivity, and availability of embryos at desired stages (although © 2014 Wiley Periodicals, Inc.
197
198
EVOLUTION & DEVELOPMENT
Vol. 16, No. 4, July–August 2014
to generating the morphological diversity of cranium and teeth among vertebrates. There are numerous reports and reviews describing the regulatory mechanisms of the NC induction, epithelial– mesenchymal transition (EMT), migration, and fate specification (For reviews, see Sakai and Wakamatsu 2005; Betancur et al. 2010; Stuhlmiller and García‐Castro 2012; Milet and Monsoro‐Burq 2012; Theveneau and Mayor 2012; McKeown et al. 2013). As for NC induction, several models have been proposed, and the latest is as follows (see also reviews mentioned above). Initially a non‐neural/neural border region in the ectoderm is specified under the influence of Wnt, FGF, and BMP signaling, by inducing the expression of “border specifier” genes such as Pax7 (e.g., Basch et al. 2006). Subsequently, appropriate levels of Wnt and BMP signalings further induce the expression of “neural crest specifier” transcription factor genes, such as Sox9/Sox10, Snail1/Snail12, at the border region (e.g., García‐Castro et al. 2002; Endo et al. 2002; Sakai et al. 2005). These transcription factors in turn control the expression of downstream genes involved in the cell–cell adhesion, cell– extracellular matrix adhesion, cytoskeletal regulation, as well as differentiation (e.g., Cheung et al. 2005; Sakai et al. 2006; Taneyhill et al. 2007; Suzuki et al. 2010; Krispin et al. 2010). In this study, to obtain insights into how the heterochronic development of the marsupial CNC cells is regulated, we examined expression patterns of genes involved in CNC formation in M. domestica embryos, and compared them with those in other amniotes. Among the genes examined, Sox9 was expressed early in a relatively broad area of the ectoderm, likely corresponding to the prospective CNC domain, and the expression persisted longer in the CNC cells during the development of the cranial anlagen, compared to that in mouse, chicken, and gecko. As the function of Sox9 is important for NC formation, EMT, and subsequent differentiation of ectomesenchymal components in the developing CNC cells (Mori‐ Akiyama et al. 2003; Akiyama et al. 2004; Sakai et al. 2006. See also Sakai and Wakamatsu 2005; Milet and Monsoro‐ Burq 2012; Theveneau and Mayor 2012 for reviews), we suggest that the spatio‐temporal alterations of Sox9 expression in the CNC cells may be the key for the heterochronic development of cranial elements of marsupials. We also show an altered expression sequence of Sox9 and Pax7, a border specifier gene. Our results thus provide a first clue to understand the molecular basis of the heterochronic development of CNC cells and jaw elements in marsupials.
MATERIALS AND METHODS
B6 mice were purchased from CLEA Japan, and midday of the vaginal plug was designated as embryonic day 0.5 (E0.5). Mouse embryos were staged according to Theiler’s stage (Theiler 1972). The breeding colony of gray short‐tailed opossum, M. domestica, is maintained at Nihon University School of Dentistry at Matsudo, under the approval of Nihon University Animal Care and Use Committee (No. AP09MD023, AP12MD015). M. domestica embryos were obtained from pregnant females, with mating video‐recorded and visually confirmed. M. domestica embryos were staged according to Mate et al. (1994), Vaglia and Smith (2003), and Smith (2006), which are adopted from McCrady stages (1938) established for Didelphis virginiana embryos. Fertilized Madagascar ground gecko (Paroedura pictus) eggs were obtained from a colony maintained at Kyoto Prefectural University of Medicine, and embryos were staged according to Noro et al. (2009). The Committee for Animal Experiment of Tohoku University Graduate School of Medicine approved the experimental procedures in this study.
In situ hybridization Whole‐mount and section in situ hybridizations were performed as described previously (Wakamatsu and Weston 1997). The DNA fragments of M. domestica Sox9, Snail1, and Snail2 corresponding to the exons were PCR‐amplified from genomic DNA prepared from M. domestica adult liver. The cDNA fragments of M. domestica BMP2, BMP4, and Pax7 were PCR‐amplified from oligo (dT) primed neonatal brain cDNA pool. These sequences were subcloned into pBluescriptII (Agilent Technologies, Santa Clara, CA, USA). The identities of M. domestica genes were confirmed by DNA sequencing. The sequences of primers were; Sox9F: GGAAGACACTGGTGATCTCC, Sox9R: CTACTCAGGGACTCTAGCTC, Snail1F: ATCCTCAGGCACAGCAGATG, Snail1R: CTAGTGTATGTCTAGTGTAC, Snail2F: CCCCATATCTCTATGAAAGC, Snail2R: AATGTGTCCTTGAAGCAACC, BMP2F: TTTTCGGGAGCAGGTGCAGG, BMP2R: ACCCACATCCCTCCACAACC, BMP4F: GGGACCAGAGAAAACCCTGC, BMP4R: CCTTCTACCACCATCTCCTG, Pax7F: TAAGCAGGCAGGAGCCAATC, Pax7R: CAACTTCATCCGACGCTGTG. cDNAs of avian Sox9 (Sakai et al. 2006), Sox10 (Cheng et al. 2000), Snail2 (Endo et al. 2002), mouse Snail1 (Sefton et al. 1998), and Sox10 (Sakai et al. 2005) for cRNA probes were described previously. Chicken Pax7 (Matsunaga et al. 2001) and mouse Sox9 cDNAs were kind gifts from Drs. Harukazu Nakamura and Ryohei Sekido, respectively. Gecko Sox9 cDNA was kindly provided by Drs. Miyuki Noro and Koji Tamura at Tohoku University (unpublished).
Experimental animals Chicken (Gallus gallus domesticus) eggs were obtained from a local farm. Embryos were staged according to Hamburger and Hamilton (HH stage, Hamburger and Hamilton 1951). Pregnant
Antibodies and immunostaining Anti‐Sox10 goat polyclonal antibody was commercially obtained (R&D systems). HNK1 (mouse IgM, Tucker et al.
Wakamatsu et al.
1988) was used as described previously (Wakamatsu et al. 2004). Fluorochrome‐conjugated secondary antibodies were purchased from Jackson Immuno Research. Immunological staining on cryo‐sections was performed as described previously (Wakamatsu et al. 1993, 1997). Sections treated with the antibody were also exposed to DAPI (Sigma‐ Aldrich, St. Louis, MO, USA) to visualize nuclei.
RESULTS AND DISCUSSION
Embryonic stages chosen for this study M. domestica embryos used in this study were staged according to previously published papers (Mate et al. 1994; Vaglia and Smith 2003; Smith 2006). Stage 18–19 embryos appeared as flat neurulae with a small area of anterior neural plate with a long primitive streak, morphologically corresponding to stage 3–4 chicken embryos. Stage 20 M. domestica embryos had anteriorly elongated neural plate with a sign of head neural fold elevation with no somites (Fig. 1a). From stage 21 to 22, the elevation of the head neural folds became apparent, and the anterior neural plate expanded laterally. In stage 22 embryos, at the lateral edges of the head neural folds, preotic and otic sulci were apparent, and a few somite pairs were formed. Embryos of this stage also showed apparent anterior–posterior elongation, and the lateral expansion of the anterior neural plate likely corresponded to the presumptive CNC region. At stage 23, a mass of delaminated CNC cells was observed (Fig. 1b, see also Smith 2006). At this stage, in addition to preotic and otic sulci, an additional notch was observed more anteriorly (designated as “anterior neural sulcus” in this study). From stage 24 to 26, the maxillo‐ mandibular and hyoid arches gradually became evident (Fig. 1c). At this stage, the elevation of neural folds was also apparent. By stage 28–29, cranial anlagen, such as fronto‐nasal mass, maxillary process, mandibular arch, and hyoid arch, as well as
Sox9 in marsupial neural crest
199
the olfactory placode, have enlarged (Fig. 1d). At these stages, the neural folds are dorsally fused along the antero‐posterior neural axis. Large fore limb buds and small hind limb buds were also observed.
Expression patterns of Snail1 in M. domestica Snail family transcription factors have been known to promote EMT of a broad range of cell types including NC. In chicken and quail embryos, Snail2 is expressed both in the CNC and trunk NC, and the expression begins prior to EMT, while Snail1 is not expressed in the NC (Sefton et al. 1998; Sakai et al. 2006). Snail2 has been known to regulate cell adhesion (Cheung and Briscoe 2003; Cheung et al. 2005; Taneyhill et al. 2007), and both blocking Snail2 activity by antisense RNA or by a dominant‐negative Snail2 inhibit EMT of CNC in vivo and in vitro, respectively (Nieto et al. 1994; Sakai et al. 2006). In mouse embryos, both Snail1 and Snail2 are expressed in migrating NC, but only Snail1 appears to be expressed prior to the EMT (Sefton et al. 1998; Sakai et al. 2005). Thus, we have cloned M. domestica homologs of both Snail1 and Snail2, and examined their expression. Snail2 expression was not observed in the CNC around the time of delamination (data not shown), hence we focused on Snail1 expression. At embryonic stage 20, Snail1 expression was strongly observed in the primitive streak (Fig. 2a), as well as in the mesodermal cells surrounding the neural plate (Fig. 2a), but no Snail1 expression was detected in the ectodermal components including the prospective CNC area. At stage 23, a bilateral expression domain of Snail1 was observed, likely corresponding to the CNC cells (Fig. 2b). Sections revealed Snail1 expression both in the CNC cells and in the mesodermal components (Fig. 2, c and d). At stage 26, when the cranial anlagen became evident, Snail1 expression was clearly observed in the CNC cells of the maxillo‐ mandibular and hyoid arches (Fig. 2, e and f), and this expression
Fig. 1. Morphologies of gray short‐tailed opossum (Monodelphis domestica) embryos examined in this study. (a) Stage 20 embryo, showing anterior neural plate (np), node (nd), and primitive streak (ps). (b) A dorsal view of stage 23 embryo. Seven pairs of somite (so) can be observed. While anterior neural plate is wide open, the CNC cells have already started generating cranial anlagen (arrowheads). Anterior neural sulcus (ans), preotic sulcus (pos), and otic sulcus (os) can be morphologically identified. (c) A dorsal view of stage 26 embryo. The cranial and trunk neural folds have not been fused yet. The maxillo‐mandibular (mx‐mb) and hyoid (hy) arches are apparent. (d) A lateral view of stage 29 embryo. The fronto‐nasal (fn), maxillary (mx), mandibular (md), hyoid (hy) processes are evident. Huge fore limb bud (fl) and small hind limb bud (hl) are also observed. Scale bars: a, 400 mm; b and c, 200 mm; d, 1 mm.
200
EVOLUTION & DEVELOPMENT
Vol. 16, No. 4, July–August 2014
Early, broad, and persistent expression patterns of Sox9 in M. domestica
Fig. 2. Expression of Snail1 mRNA in M. domestica embryos. (a) Dorsal view of a stage 20 embryo. While strong expression of Snail1 in the primitive streak (ps) and a weaker expression in the lateral mesoderm are observed, little expression is detectable in the CNC domain (arrow). (b) A dorsal view of stage 23 embryo. The expression of Snail1 in the CNC cells is clearly observed (arrowhead). The expression in the somites (so) is also detected. The planes of sections shown in (c) and (d) are indicated. (c) A transverse section of the embryo shown in (b). Snail1 expression is evident in the CNC cells (arrowheads) and paraxial mesoderm (mes), but not in the neural plate (np). (d) A transverse section of the embryo shown in (b). Snail1 expression is detected in the somite (so). (e) A dorsal view of stage 26 embryo. The cranial processes such as maxillo‐mandibular (mx‐md) and hyoid (hy) arches strongly express Snail1. In addition to the expression in the somites and primitive streak, the lateral plate mesoderm corresponding to the future fore limb bud (fl) also expresses Snail1. (f) A lateral view of embryo shown in (e). Snail1 expression in maxillo‐mandibular (mx‐ md) and hyoid (hy) arches is evident. The anterior neural sulcus (see also Fig. 1b) is indicated by an arrowhead. (g) A section of stage 29 embryo at the hindbrain level. Snail1 is expressed homogenously in the CNC cell‐derived ectomesenchyme of the mandibular arch (md). Scale bars: a, 400 mm; b and e, 200 mm.
continued at least until stage 29 (Fig. 2g). We also noted that mesenchymal cells rostral to the anterior neural sulcus were negative for Snail1 expression (Fig. 2f). Because these cells expressed Sox9 (see below), and because Snail1 is expressed in the maxillio‐mandibular CNC cells but not in the fronto‐nasal CNC cells of mouse embryos (Sefton et al. 1998), these cells likely corresponded to the fronto‐nasal population of CNC cells of M. domestica embryos. Snail1 expression in the fore limb field and the paraxial mesoderm was also similar to that in mouse embryos (Sefton et al. 1998. See also Keyte and Smith 2010 for M. domestica limb development), thus the expression pattern of Snail1 seemed to be largely conserved in M. domestica and mouse.
Group E Sox transcription factor genes, such as Sox9 and Sox10, are important for the promotion of EMT and differentiation of NC (See reviews mentioned above). In particular, Sox9 has been shown to be crucial for the reduction of cell–cell adhesion during EMT of NC, the growth of cranial processes, and the differentiation of cartilage in mouse and avian systems (Lefebvre et al. 1997; Mori‐Akiyama et al. 2003; Akiyama et al. 2004; Suzuki et al. 2006; Sakai et al. 2006). We thus examined the expression of Sox9 in M. domestica embryos. At stage 18, no bilateral expression of Sox9 was observed (Fig. 3a). The onset of Sox9 expression in the prospective CNC was, therefore, around stage 19 (Fig. 3b), prior to neural fold elevation, and well prior to segmentation of the first somite. Observation on whole‐mount and sections revealed that, at stage 20, the Sox9‐positive cells still remained as a part of the ectodermal epithelium prior to EMT (Fig. 3, c and d). At stage 23, many Sox9‐positive CNC cells have delaminated, and consisted of virtually all the mesenchyme cells of the fronto‐ nasal and maxillo‐mandibular processes (Fig. 3, e and f). At stage 25, Sox9‐positive cells still occupied the mesenchyme of the cranial processes including the hyoid arches (Fig. 3, g and h). Unlike Snail1, Sox9‐positive CNC cells were clearly found anteriorly to the anterior neural sulcus, consistent with these cells being the fronto‐nasal CNC cells (Fig. 3h). At stage 28, Sox9 expression was still observed in the fronto‐nasal, maxillary, mandibular processes, as well as hyoid arches (Fig. 3i). At stage 29, Sox9 expression was largely lost in the fronto‐ nasal, mandibular and hyoid, but remained in the maxillary processes (Fig. 3, j, k, l). From stage 20–29, we noted the expression of Sox9 in more posterior NC including trunk NC (Fig. 3, c, e, g, j). We also examined Sox10 protein expression in M. domestica by immunostaining (Fig. 4). At stage 20, The CNC cells forming the cranial anlagen strongly expressed Sox10 (Fig. 4, a–a00 ). At stage 26, the CNC‐derived ectomesenchymal cells in the cranial processes such as the mandibular arch expressed Sox10 protein, but the expression level was lower compared to the more dorsally positioned CNC cells (Fig. 4, b–b00 ). Based on their location, these dorsal CNC cells were likely to be progenitors of the cranial ganglia (see below). At stage 28, Sox10 expression was barely detectable in the CNC‐derived ectomesenhcyme of the cranial anlagen, while high Sox10 expression was clearly observed in the cranial ganglia (Fig. 4, c–c00 ). These observations indicated that CNC cells co‐expressed Sox9 and Sox10 for a long period, but Sox10 expression diminished earlier than Sox9 in the cranial ectomesenchyme. The persistent expression of Sox9 in CNC cells and co‐ expression of Sox9 and Sox10 in the CNC‐derived ectomesenchyme of the cranial processes in M. domestica seemed to be different from those in avian embryos. We have previously
Wakamatsu et al.
Sox9 in marsupial neural crest
201
Fig. 3. Expression of Sox9 mRNA in NC cells of M. domestica embryos. (a–c) A series of M. domestica embryos prior to the CNC cell delamination. At stage 18, Sox9 expression is weakly and broadly detected in the neural plate (a). At stage 19, Sox9 expression is upregulated bilaterally in the presumptive CNC domain (b, arrowheads). At stage 20, Sox9 expression in the CNC is evident (c, arrowhead). Premigratory NC cells in the more posterior region are also recognized by Sox9 expression (arrows). (d) A transverse section of the embryo shown in (c) at the indicated level. Sox9 expression in the premigratory CNC (arrowheads) is observed bilaterally to the neural plate (np). (e) A dorsal view of stage 23 embryo, showing Sox9 expression both in the CNC (arrowhead) and the trunk NC (arrows). Sox9 expression is also observed in the somites (so). The preotic sulcus (pos) is indicated. The axial level corresponding to sections shown in (f) is also indicated. (f) A transverse section of a stage 23 embryo similar to the embryo in (e). Sox9 mRNA is expressed in the CNC cells forming the maxillo‐mandibular anlagen (arrowheads), as well as in the mesodermal cells (mes). (g and h) Dorsal (g) and lateral (h) views of a stage 25 embryo. Sox9 expression is evident in the CNC cells of fronto‐nasal (fn), maxillo‐mandibular (mx‐md), and hyoid (hy) processes. Sox9 expression is also detected in the otic vesicle (ov), somites, and trunk NC (arrows in g). Some premigratory CNC cells remain at the lateral edge of the head neural fold (arrows in h). Anterior neural sulcus is indicated (arrowhead in h). (i) Side view of the cranial region of a stage 28 embryo. Sox9 is expressed in the fronto‐nasal (fn), maxillary (mx), mandibular (md), and hyoid (hy) processes. (j) Side view of a stage 29 embryo. Trunk NC cells express Sox9 (arrows). fl: fore limb bud. (k) High magnification view of the head of the embryo, shown in (j). The expression of Sox9 is significantly reduced in the cranial anlagen except for the maxillary process (mx). (l) Section of stage 29 embryo at the level of the hindbrain, showing Sox9 expression. Note the lack of Sox9 mRNA in the distal‐medial region of the mandibular arch (md). Scale bars: a–c, e, g, 200 mm; i–k, 1 mm.
shown that, in chicken and quail embryos, Sox9 expression in the head neural folds was first observed at stage 7 (1 somite), and the expression was up‐regulated in the premigratory CNC cells by stage 8 (4 somites), but Sox9 expression was quickly down‐ regulated at the early phase of CNC migration (see Suzuki et al. 2006; Sakai et al. 2006). This down‐regulation coincided with Sox10 expression, probably because Sox10 could repress Sox9, while Sox9 could induce Sox10 (Suzuki et al. 2006). Because detailed observation of Sox9/Sox10 expression has not been done in the CNC of eutherian mammals, we also performed whole‐mount in situs of mouse embryos from E8.0 to E8.75 (Theiler stage 11d‐13, Fig. 5) for further comparisons. Neither Sox9 nor Sox10 was expressed in the neural folds of the pre‐ somitegenesis stage embryos at E8.0 (Theiler stage 11d, Fig. 5, a and f), and Sox9 expression was first turned on in the premigratory CNC of the cranial neural folds around E8.0‐
8.25 (Theiler stage 12a, Fig. 5b). Thus, the onset of Sox9 expression in chicken and mouse was relatively later than in M. domestica if somitegenesis was used as a standard of developmental stage. Sox9 expression was evident in the mouse CNC cells at the early phase of migration (Theiler stage 12a, Fig. 5c), but diminished during migration (Theiler stage 12b, Fig. 5d), and was turned off in the CNC cells entering the cranial processes at 8.75 dpc (Theiler stage 13, Fig. 5e). In mouse embryos at similar stages, Sox10 was weakly detected in the premigratory CNC cells at E8.25 (Theiler stage 12a, Fig. 5g), was upregulated in the migrating CNC cells at E8.25–8.5 (Theiler stage 12a–12b, Fig. 5, h and i), and persisted in the CNC cells forming the cranial processes at E8.5 (Theiler stage 13, Fig. 5j). Thus, the transition of the group E Sox genes from Sox9 to Sox10 in the mouse CNC cells resembled to that in chicken, but not to that in M. domestica.
202
EVOLUTION & DEVELOPMENT
Vol. 16, No. 4, July–August 2014
Fig. 4. Expression of Sox10 protein in CNC cells of M. domestica embryos. Transverse sections of stage 23 (a–a00 ), stage 26 (b–b00 ), and stage 28 (c–c00 ) embryos at the level of the mandibular arch are shown. DAPI staining shows the nuclei. (a–a00 ) At stage 23, the CNC cells strongly express Sox10 protein (arrowheads), but not in other tissues, such as neural plate (np) and mesoderm (mes). (b–b00 ) At stage 26, Sox10 expression is detected in the CNC‐derived ectomesenchyme (arrowheads) of the mandibular arch (md). Note that the staining intensity in the ectomesenchyme is lower than that in the late migrating CNC cells remaining dorsally ( ). (c–c00 ) At stage 28, while sensory ganglia strongly express Sox10 protein ( ), the expression is barely detectable in the mandibular ectomesenchyme (arrowheads).
It is possible that the persistent expression of Sox9 in M. domestica might be a unique feature acquired in the marsupial lineage. Alternatively, the persistent expression of Sox9 might be an ancestral feature for mammals, and that the similar sequential expression of Sox9/Sox10 in chicken and mouse might have been established independently, as a result of an evolutionary convergence. To explore this further, we examined the expression of Sox9 in gecko embryos, as an example of a reptile (Fig. 6). At 0.5 dpo (days postoviposition), Sox9 expression was evident along the dorsal midline of the midbrain as the left and right neural folds were fused (Fig. 6, a and b). The CNC identity of these Sox9‐positive cells was confirmed by HNK1 staining on sections (Fig. 6, b–g). Thus, the late onset of the CNC cell migration at the time of cranial neural fold closure appears similar in gecko and chicken embryos, as well as in crocodile embryos as previously reported (Kundrat 2009), but not in mouse or M. domestica embryos, in which the CNC cells started migrating well before the fusion of the head neural folds (Figs. 3, e, f, and 5c). A stream of Sox9 expression from the caudal end of the dorsal midline of the midbrain was also observed in gecko embryos (Fig. 6, a, e, f, h, i), suggesting that,
Fig. 5. Expression of Sox9 and Sox10 mRNA in CNC cells of mouse embryos. The mRNA expression of Sox9 (a–e) and Sox10 (f–j) are shown from E8 to E8.75 (Theiler stage 11d‐13). An intense Sox9 expression is detected earlier than Sox10 expression in the premigratory CNC along the lateral edge of the cranial neural folds (arrowheads; compare (a, b) and (f, g)). Sox9 expression in the CNC cells is downregulated during migration (d), while Sox10 expression persists (i). At the oldest stage shown in this figure, Sox9 expression is largely absent in the cranial processes (e), although Sox9 expression is detectable in the otic placode (op) and the trunk NC cells (arrows in e). At similar stage, Sox10 expression is observed in all the cranial processes, such as fronto‐nasal (fn), mandibular (md), and hyoid (hy), as well as in the trunk NC cells (arrows in j). Scale bars: a and f, 100 mm; b–e and g–j, 200 mm.
Fig. 6. Expression of Sox9 mRNA in CNC cells of gecko embryos. (a) Side view of 0.5 days postoviposition (dpo) gecko embryo, showing Sox9 mRNA expression in the CNC cells (arrowheads) and otic placode (op). A large arrowhead indicates CNC cells on top of the fore‐midbrain. Small arrowheads indicate posterior migratory stream of the midbrain CNC cells. (b–d) A section of a Sox9 in situ hybridized embryo shown in (a), counter‐stained with HNK1 antibody and DAPI. The plane of section is indicated in (a). HNK1‐ positive CNC cells reside on top of the midbrain (large arrowhead). (e–g) A section of (a), counter‐stained with HNK1 antibody and DAPI. The plane of section is indicated in (a). Most CNC cells in the posterior migratory stream are Sox9 and HNK1‐double positive (small arrowhead), although the Sox9‐positive CNC cells residing in the most distal part of the stream appears to be HNK1‐negative. (h) A side view of the head of 0.5 dpo embryo (slightly older than the embryo shown in (a). Only a few Sox9‐positive CNC cells remain on the top of midbrain (an arrow), and many Sox9‐positive CNC cells occur in the posterior migratory stream (arrowheads). (i) Side view of the head of 1.5 dpo embryo. While Sox9 expression remains in the posterior migratory stream (arrowheads), Sox9 expression is absent in the maxillo‐mandibular arches (mx‐md). Scale bars: a, 2 mm; b and c, 1 mm.
Wakamatsu et al.
Fig. 7. Expression of BMP2/4 mRNA in chicken, mouse, and M. domestica embryos at stages prior to CNC cell migration. (a–c) BMP4 mRNA expression in chicken embryos at stage 5 (a), stage 6 (b), and stage 8.5 (c). At stage 5, BMP4 expression is detected in the non‐neural ectoderm surrounding the neural plate (arrowheads in a). At later stages, BMP4 expression in the non‐neural ectoderm is downregulated, while strong expression along the neural folds is observed (arrows in b, c). BMP4 is also expressed in the primitive streak (ps) of chicken embryos (a–c). (d) BMP2 expression in the non‐neural ectoderm surrounding the neural plate of E8.0 (Theiler stage 11d) mouse embryo (dorsal view, arrowheads). (e–h) BMP2 mRNA expression in M. domestica embryos at stage 19 (e), stage 20 (f), stage 21 (g), and stage 22 (h). BMP2 expression is observed in the non‐neural ectoderm surrounding the neural plate (arrowheads) and in the cranial neural folds of older embryos (arrows). Scale bars: a–d, 100 mm; e–h, 200 mm.
unlike the CNC cells of chicken, mouse, or M. domestica, in which the midbrain CNC cells migrate broadly toward the position of the future maxillo‐mandibular anlage, the gecko midbrain CNC cells appeared to have a more restricted migratory pathway. This posterior migratory stream of the midbrain CNC has been previously identified in crocodile embryos (Kundrat 2009). It seemed likely that the posterior migratory pathway of the CNC cells found in lepidosaurian gecko and archosaurian crocodile represents the ancestral state of reptilia, and that the broad migratory stream of midbrain CNC cells in mammals and archosaurian chicken would be the result of the evolutionary convergence. Nevertheless, in 0.5 and 1.5 dpo gecko embryos, Sox9 expression was not found more ventrally in the cranial anlagen (Fig. 6, h and i), suggesting that, in the gecko CNC cells, the Sox9 expression was transient, similar to that in chicken and mouse embryos, but not to that in M. domestica. Thus, the persistent expression of Sox9 in the ectomesenchymal
Sox9 in marsupial neural crest
203
Fig. 8. Pax7 expression in chicken (a–e) and M. domestica (f–i) embryos. (a–e) From stage 5 to 7, Pax7 expression is observed along the neural border of chicken embryos (arrows in a–d). At stage 7, the anterior part of the Pax7‐positve border domain slightly expands (d). At stage 8.5, Pax7 is strongly expressed in CNC cells at the onset of their migration (arrows in e). The midbrain‐hindbrain boundary (mhb) is also indicated. (f) At the earliest stage of M. domestica embryo examined (stage 18), Pax7 expression is only faintly observed in the neural plate. (g) At stage 20, Pax7 expression is detectable along the border region (arrows). (h) At stage 21.5, while Pax7 expression along the border mostly remains as a lateral thin line (arrows), in the premigratory CNC region, the Pax7 expression domain expands medially (arrowheads). Preotic sulcus (pos) is indicated. (i) A high magnification view of the anterior part of the embryo in (h) showing lateral high‐medial low gradient of Pax7 expression in the premigratory CNC domain (arrowheads). Arrows indicate the Pax7 expression along the lateral edge of the neural fold. Scale bars: a–e, 100 mm; f–h, 200 mm.
CNC cells of M. domestica embryos seemed to be a peculiarity of development of the cranial anlagen in Marsupials rather than an ancestral feature.
Expression patterns of BMPs in M. domestica BMP signaling is essential for CNC induction in various amniotes (Kanzler et al. 2000; Endo et al. 2002; see also reviews listed above). In chicken embryos, BMP4 was initially expressed in the non‐neural ectoderm at stage 5, well prior to the onset of Sox9 expression in the premigratory CNC (Fig. 7a), and as the non‐neural ectoderm expression gradually diminished, the cranial neural folds expressed BMP4 at stage 7 and 8.5 (Fig. 7, b and c), coincided with the elevation of Sox9 expression in the premigratory CNC cells (see above). In mouse embryos, BMP2 was expressed in the non‐neural ectoderm at E8.0 (Theiler stage 11d, Fig. 7d; see also Kanzler et al. 2000), prior to the onset of Sox9 expression in the premigratory CNC. Thus, we have cloned the M. domestica homologs of BMP2 and BMP4, and
204
EVOLUTION & DEVELOPMENT
Vol. 16, No. 4, July–August 2014
Table 1. Comparisons of Pax7 expression, Sox9 expression, and the migratory behavior of CNC cells in mouse, opossum, gecko, and chicken embryos
Pax7 expression
The onset of Sox9 expression
The diminishing point of Sox9 expression
The onset of CNC cell migration
Migratory behavior of midbrain CNC cells
Mouse
Neural folds (Basch et al. 2006)
Later than the first somite stage, at the middle phase of neural fold elevation (this study)
Early: during migration (this study)
Middle phase of neural fold elevation (this study)
Anterior‐posteriorly unrestricted (Osumi‐ Yamashita et al. 1997)
Opossum
Initially surrounding the anterior neural plate, then bilateral bands (this study)
Late: in the cranial processes (this study)
n.d.
Chicken (and quail)
Bilateral bands (Basch et al. 2006, this study)
At the onset of somitegenesis, at the middle phase of neural fold elevation (Sakai et al. 2006)
Early: during migration (Suzuki et al. 2006)
Beginning of neural fold elevation (This study. See also Vaglia and Smith 2003) Neural fold closure (This study. See Kundrat 2009, for crocodile) Neural fold closure (This study. See also Kundrat 2009, for ostrich)
Anterior‐posteriorly unrestricted (Vaglia and Smith 2003, this study)
Gecko
Before the first somite stage, at the beginning of neural fold elevation (this study) n.d.
Early: during migration (this study)
Posterior migratory stream (This study. See Kundrat 2009, for crocodile) Anterior‐posteriorly unrestricted (see Le Douarin and Kalcheim 1999)
n.d., Not determined.
examined their expression patterns (Fig. 7, e–h). In M. domestica embryos, BMP4 expression was not detected in the stage 20–23 embryos, although at the trunk level of stage 29 embryos, dorsal neural tube expression was observed, similar to that in mouse and chicken embryos (data not shown). In contrast, M. domestica BMP2 was detected in the non‐neural ectoderm at stage 19 (Fig. 7e). In older embryos such as stage 20, 21, and 22, while BMP2 expression in the non‐neural ectoderm gradually regressed peripherally (Fig. 7, f–h), BMP2 expression along the lateral edge of the neural folds became evident (Fig. 7, f–h).
Because this expression pattern of BMP2 in M. domestica embryos appeared to be similar to BMP4 in chicken embryos, BMP2, but not BMP4, might be involved in CNC formation of M. domestica.
Expression patterns of Pax7 in M. domestica Previous studies in chicken, frog, and zebrafish embryos suggested that, prior to the induction of “neural crest specifier” genes such as Snail1/Snail2 and Sox9, the non‐neural/neural
Fig. 9. Diagrammatic comparisons of morphological events, developmental status of the CNC, and Sox9/Sox10 expression in the CNC among chicken, mouse and opossum. Developmental stages are also indicated. Note the early onset of the Sox9 expression, the cranial anlagen protrusion, and the late fusion of neural fold in the opossum. (Note the onset of the Sox10 expression in the opossum is not determined.)
Wakamatsu et al.
border has been established by a combination of inductive signals (Milet and Monsoro‐Burq 2012, for a review). In chicken embryos, Pax7 expression was a good indicator of border specification, and Pax7 function was required for the expression of NC specifier genes, including Sox9 (Basch et al. 2006). Thus, we compared Pax7 expression in chicken (Fig. 8, a–e) and M. domestica embryos (Fig. 8, f–i). At stage 18 of M. domestica embryo, prior to Sox9 expression, Pax7 expression was not detected at the border along the newly formed neural plate (Fig. 8f). At stage 20, Pax7 expression was observed in the neural border (Fig. 8g) in a very similar manner to that in stage 4– 7 chicken embryos (Fig. 8, a–c). However, this expression pattern was different from the broad, bilateral expression domain of Sox9 in the prospective CNC at similar stages of M. domestica embryos (see Fig. 3, b and c). In addition, the onset of Pax7 expression in the border was no earlier than Sox9 expression in the premigratory CNC (compare Figs. 3, a–c and 8, f and g). Thus, the inductive sequence of the non‐neural/neural border specification demarcated by Pax7 expression and the subsequent CNC induction indicated by the bilateral expression of Sox9 found in chicken is difficult to apply to M. domestica embryos. At older stages of M. domestica embryos such as stage 21.5, a lateral expansion of the Pax7 domain was observed in the area corresponding to the premigratory CNC expressing Sox9 (Fig. 8, h and i; see also Fig. 3). Interestingly, this Pax7 expression showed a “lateral high‐medial low” gradient (Fig. 8i). We thus speculate that Pax7 expression in CNC would be under the influence of medio‐lateral graded signals, such as BMP2 expressed at the lateral edge of the neural folds. Nonetheless, the distinct expression sequence of Pax7 and Sox9 in M. domestica suggested the altered epistatic relationship between these genes, and its contribution to heterochronic formation of the CNC.
CONCLUSION In this study, we examined gene expression patterns in the neural plate and CNC cells of M. domestica embryos, and correlated them with the early developmental events and the process of the CNC development. Our results are summarized in Table 1 and Fig. 9. Sox9 expression in M. domestica CNC cells is apparently modified, compared to other amniotes. The early and broad expression of Sox9 in the premigratory CNC domain clearly corresponds to, and likely contributes to, the early generation of many CNC cells shown in the previous histological studies (Smith 2001; Vaglia and Smith 2003; Smith 2006). Although the role of the persistent expression of Sox9 in the CNC cell‐derived ectomesenchyme of cranial processes remains unclear, it may also contribute to the cartilage differentiation of the ectomesenchyme. This study also indicated that, because gecko (this study) and crocodile (Kundrat 2009) CNC cells at the midbrain level have a conserved posterior migratory stream, the lack of this
Sox9 in marsupial neural crest
205
migratory stream found in the avian and mammalian embryos might have evolved independently. Acknowledgments We thank Drs. Koji Tamura and Don Newgreen for comments on the manuscript. We thank Drs. Harukazu Nakamura, Ryohei Sekido, Koji Tamura, and Miyuki Noro for plasmids. We also thank Noriko Uemura for opossum care. This work was supported in part by a grant to Y.W. from JSPS (KAKENNHI Grant Number 24570228), and to K.S. (KAKENNHI Grant Number 20592154).
REFERENCES Akiyama, H., et al. 2004. Essential role of Sox9 in the pathway that controls formation of cardiac valves and septa. Proc. Natl. Acad. Sci. 101: 6502– 6507. Basch, M. L., Bronner‐Fraser, M., and Garcia‐Castro, M. I. 2006. Specification of the neural crest occurs during gastrulation and requires Pax7. Nature 441: 218–222. Betancur, P., Bronner‐Fraser, M., and Sauka‐Spengler, T. 2010. Assembling neural crest regulatory circuits into a gene regulatory network. Annu. Rev. Cell. Dev. Biol. 26: 581–603. Cheng, U., Cheung, M., Abu‐Elmagd, M. M., Orme, A., and Scotting, P. J. 2000. Chick sox10, a transcription factor expressed in both early neural crest cells and central nervous system. Brain Res. Dev. Brain Res. 121: 233–241. Cheung, A. F. P., et al. 2009. The subventricular zone is the developmental milestone of a 6‐layered neocortex: comparisons in metatherian and eutherians mammals. Cereb. Cortex 20: 1071–1081. Cheung, M., and Briscoe, J. 2003. Neural crest development is regulated by the transcription factor Sox9. Development 130: 5681–5693. Cheung, M., Chaboissier, M. C., Mynett, A., Hirst, E., Schedl, A., and Briscoe, J. 2005. The transcriptional control of trunk neural crest induction, survival, and delamination. Dev. Cell. 8: 179–192. Debora, C. K., and Sears, K. E. 2010. The divergent development of the apical ectodermal ridge in the marsupial Monodelphis domestica. Anat. Rec. 290: 1325–1332. Endo, Y., Osumi, N., and Wakamatsu, Y. 2002. Bimodal functions of Notch‐ mediated signaling are involved in neural crest formation during avian ectoderm development. Development 129: 863–873. García‐Castro, M., Marcelle, C., and Bronner‐Fraser, M. 2002. Ectodermal Wnt function as a neural crest inducer. Science 297: 848–851. Hall, B. K. 2000. The neural crest as a fourth germ layer and vertebrates as quadroblastic not triploblastic. Evol. Dev. 2: 3–5. Hamburger, V., and Hamilton, H. L. 1951. A series of normal stages in the development of the chick embryo. J. Morph. 88: 49–92. Kanzler, B., Foreman, R. K., Labosky, P. A., and Mallo, M. 2000. BMP signaling is essential for development of skeletogenic and neurogenic cranial neural crest. Development 127: 1095–1104. Keyte, A. L., and Smith, K. K. 2010. Developmental origins of precocial forelimbs in marsupial neonates. Development 137: 4283–4294. Krispin, S., Nitzan, E., Kassam, Y., and Kalcheim, C. 2010. Evidence for a dynamic spatiotemporal fate map and early fate restrictions of migratory avian neural crest. Development 137: 585–595. Kundrat, M. 2009. Heterochronic shift between early organogenesis and migration of cephalic neural crest cells in two divergent evolutionary phenotypes of archosaurs: crocodile and ostrich. Evol. Dev. 11: 535–546. Le Douarin, N. M., and Kalcheim, C. 1999. The Neural Crest. 2nd Ed. Cambridge University Press, Cambridge. Lefebvre, V., Huang, W., Harley, V. R., Goodfellow, P. N., and de Cromburgghe, B. 1997. SOX9 is a potent activator of the chondrocyte‐ specific enhancer of the proa1(II) collagen gene. Mol. Cell. Biol. 17: 2336–2346. Mate, K. E., Robinson, E. S., Vandeberg, J. L., and Pedersen, R. A. 1994. Timetable of in vivo embryonic development in the gray short‐tailed opossum (Monodelphis domestica). Mol. Rep. Dev. 39: 365–374.
206
EVOLUTION & DEVELOPMENT
Vol. 16, No. 4, July–August 2014
Matsunaga, E., Araki, I., and Nakamura, H. 2001. Role of Pax3/7 in the tectum regionalization. Development 128: 4069–4077. McCrady, E. 1938. The Embryology of the Opossum. Wistar Institute of Anatomy and Biology, Philadelphia. p. 233. McKeown, S. J., Wallace, A. S., and Anderson, R. B. 2013. Expression and function of cell adhesion molecules during neural crest migration. Dev. Biol. 373: 244–257. Milet, C., and Monsoro‐Burq, A. H. 2012. Neural crest induction at the neural plate border in vertebrates. Dev. Biol. 366: 22–33. Mori‐Akiyama, Y., Akiyama, H., Rowitch, D. H., and de Crombrugghe, B. 2003. Sox9 is required for determination of the chondrogenic cell lineage in the cranial neural crest. Proc. Natl. Acad. Sci. 100: 9360–9365. Moustakas, J. E., Smith, K. K., and Hlusko, L. 2011. Evolution and development of the mammalian dentition: insights from marsupial Monodelphis domestica. Dev. Dyn. 240: 232–239. Nieto, M. A., Sargent, M. G., Wilkinson, D. G., and Cooke, J. 1994. Control of cell behavior during vertebrate development by Slug, a zinc finger gene. Science 264: 835–839. Noro, M., Uejima, A., Abe, G., Manabe, M., and Tamura, K. 2009. Normal developmental stages of the Madagascar ground Gecko Paroedura pictus with special reference to limb morphogenesis. Dev. Dyn. 238: 100–109. Osumi‐Yamashita, N., Ninomiya, Y., and Eto, K. 1997. Mammalian craniofacial embryology in vitro. Int. J. Dev. Biol. 41: 187–194. Sakai, D., Tanaka, Y., Endo, Y., Osumi, N., Okamoto, H., and Wakamatsu, Y. 2005. Regulation of Slug transcription in embryonic ectoderm by beta‐ catenin‐Lef/Tcf and BMP‐Smad signaling. Dev. Growth Differ. 47: 471– 482. Sakai, D., and Wakamatsu, Y. 2005. Regulatory mechanisms for neural crest formation. Cells Tissues Organs 179: 24–35. Sakai, D., Suzuki, T., Osumi, N., and Wakamatsu, Y. 2006. Cooperative action of Sox9, Snail2, and PKA signaling in early neural crest development. Development 133: 1323–1333. Sefton, M., Sanchez, S., and Nieto, M. A. 1998. Conserved and divergent roles for members of Snail family of transcription factors in the chick and mouse embryo. Development 125: 3111–3121. Smith, K. K. 2001. Early development of the neural plate, neural crest and facial region of marsupials. J. Anat. 199: 121–131.
Smith, K. K. 2006. Craniofacial development in marsupial mammals: developmental origins of evolutionary change. Dev. Dyn. 235: 1181– 1193. Stuhlmiller, T. J., and García‐Castro, M. I. 2012. Current perspectives of the signaling pathways directing neural crest induction. Cell. Mol. Life Sci. 69: 3715–3737. Suzuki, T., Sakai, D., Osumi, N., Wada, H., and Wakamatsu, Y. 2006. Sox genes regulate type 2 collagen expression in avian neural crest cells. Dev. Growth Differ. 48: 477–486. Suzuki, T., Osumi, N., and Wakamatsu, Y. 2010. Stabilization of ATF4 protein is required for the regulation of epithelial–mesenchymal transition of the avian neural crest. Dev. Biol. 344: 658–668. Taneyhill, L. A., Coles, E. G., and Bronner‐Fraser, M. 2007. Snail2 directly represses cadherin6B during epithelial‐to‐mesenchymal transitions of the neural crest. Development 134: 1481–1490. Theveneau, E., and Mayor, R. 2012. Neural crest delamination and migration: from epithelium‐mesenchyme transition to collective cell migration. Dev. Biol. 366: 34–54. Tucker, G. C., Delarue, M., Zada, S., Boucaut, J. C., and Thiery, J. P. 1988. Expression of the HNK‐1/NC‐1 epitope in early vertebrate neurogenesis. Cell Tissue Res. 251: 457–465. Theiler, K. 1972. The House Mouse. Springer‐Verlag, New York. Vaglia, J. L., and Smith, K. K. 2003. Early differentiation and migration of cranial neural crest in the opossum, Monodelphis domestica. Evol. Dev. 5: 121–135. Wakamatsu, Y., Watanabe, Y., Shimono, A., and Kondoh, H. 1993. Transition of localization of the N‐myc protein from nucleus to cytoplasm in differentiating neurons. Neuron 10: 1–9. Wakamatsu, Y., and Weston, J. A. 1997. Sequential expression and role of Hu RNA‐binding proteins during neurogenesis. Development 124: 3449– 3460. Wakamatsu, Y., Watanabe, Y., Nakamura, H., and Kondoh, H. 1997. Regulation of the neural crest cell fate by N‐myc: promotion of ventral migration and neuronal differentiation. Development 124: 1953–1962. Wakamatsu, Y., Endo, Y., Osumi, N., and Weston, J. A. 2004. Multiple roles of SOX2, a HMG‐box transcription factor in avian development. Dev. Dyn. 229: 74–86.
