An old modeling mechanism similar to Turing regulates the development of skin dentition in sharks



Vertebrates have a wide range of epithelial appendages, including scales, feathers and hairs. The pattern of development of these various structures can be explained theoretically by Alan Turing's reaction-diffusion system. However, the role of this system in standardizing the epithelial appendage of early divergent strains (compared to tetrapods), such as cartilaginous fish, is poorly understood. We investigated the pattern of the tooth's unique tooth-shaped denticles of sharks, which is closely related to their hydrodynamic and protective functions. We have demonstrated through simulation models that a Turing-like mechanism can explain the standardization of shark dentition and verify this system using gene expression analysis and gene pathway inhibition experiments. This mechanism bears a striking resemblance to bird feather patterns, suggesting a profound homology to the system. We propose that a diverse range of vertebrate appendages, from shark denticles to bird feathers and mammalian hair, use this ancient and conserved system, with a slight genetic modulation, responsible for wide variations in standardization.


Vertebrates have a myriad of diverse epithelial appendages, including hair, feathers, scales, spines, and teeth (1). Recent research has revealed that these structures share extensive homology of development as they grow from a common base: the epithelial placode (24). Despite this shared ancestry, there are large variations in both the final morphology and the spatial arrangement of these organs (1). Such variation in standardization has evolved to facilitate various functions, for example, trawl reduction, thermoregulation and communication (57).

Alan Turing's model of reaction-diffusion (RD) provides an explanation for the diversity of patterns observed in nature (812). This model describes how interactions between fungi diffusing through tissue can give rise to autonomous patterns of epithelial appendages (8, 13). These morphogens typically constitute two interactive molecular signals that play the role of a short-range activator and a long-range inhibitor (14). The autocatalytic activator promotes its own expression and expression of the inhibitor, which, in turn, represses the activator. Turing has shown that when properly tuned, the kinetics of the non-linear reaction and the difference in the diffusion coefficients may result in the formation of a stable periodic pattern in an initially homogeneous signal field in which the activator peaks alternate with the inhibitor (15). This self-organized system defines the spatial distribution of the placodes and, therefore, the standardization of epithelial appendages. It is worth noting that, in addition to DR, other factors such as mechanosensitization of the tissue may be important for the control of the appendix pattern of the skin (16). In this case, the standardization may still be via Turing instability, but using mechanics in addition to the molecular interactions RD (17). We refer to this as a system similar to Turing.

There is a growing body of experimental research supporting the modeling of DR throughout the development of the epithelial appendix. This includes the role of DR in the standardization and morphogenesis of feathers and hair (1821). These studies revealed that molecular signals, such as fibroblast growth factors (FGFs) and sonic hedgehog (Shh) may play autocatalytic activating functions, whereas bone morphogenetic proteins (BMPs) may act as inhibitors (18, 22). Despite the evidence for RD patterns in classical tetrapod model organisms (ie, mice and pups), our understanding of this system in earlier divergent strains is limited.

Chondrichthyans (cartilaginous fish) occupy the sister line of osteichthyans (osteous vertebrates) and constitute a divergent lineage older than tetrapods. Elasmobranchs (sharks, skates and rays) are a subclass of Chondrichthyes, which have hard and mineralized epithelial appendages, known as odontódeos. Odontodes include teeth and dermal denticles, consisting of a pulp cavity encapsulated in layers of dentin and enameloid (23). Odontogenic competence is believed to have originated in the dermal skeleton, originating the denticles as a precursor of vertebrate oral dentition (2426). These structures were observed in early vertebrates that lived 450 million years ago (27, 28). Denticles have evolved to fulfill a variety of functions, including drag reduction and shielding (5, 29). It has been previously suggested that shark denticles do not follow a strict spatial pattern (30, 31), although they exhibit intraspecific and interspecific variations in morphology and standardization, which is closely related to their function (32, 33). Recent research has suggested that a RD mechanism may be behind the arrangement of denticles in a fossil adult cretaceous shark (Tribodus limae(I.e.34). However, experimental evidence addressing the onset of standardization and its genetic basis is needed to determine the role of this system in elasmobranchs.

The concept of Reif inhibitory field is considered the main hypothesis to explain the standardization of odontódeos (35). This theory describes how diffusion from existing odontodes can dictate the proximity of contemporary units by preventing the formation of placodes within the perimeter of zones of inhibition around existing teeth or denticles (35, 36). However, no underlying molecular basis has been identified to support this idea. In fact, it was described as a verbal description of a restricted parameterization of a DR system (34).

It is believed that there is initial morphogenetic similarity between the shark's teeth and the chick plumage pattern, the latter of which is controlled by RD (18, 37). The chick feathers initially develop sequentially in a dorsal longitudinal line along the midline of the embryo. This primer line triggers the subsequent formation of placode in adjacent parallel lines until the integument is covered (3840). This is consistent with an RD system (8, 18). Embryonic sharks develop two dorsolateral rows of increased denticles that emerge before the subsequent eruption of intricate body denticles (Fig. 1) (36, 41, 42). Immediately after hatching, these lines are included in the general scale (26). As observed during standardization of feathers (18, 39), the dorsal denticles of shark can act as initiating lines that trigger the emergence of adjacent denticles, following a system similar to Turing.

Figure 1 DR modeling may explain the standardization of dentition in cats.

(AN) Catsharks exhibit two rows of dorsal dental placodes (PD) at the developmental stage 32 (~ 80 dpf). (B for AND and G for J) These placodes undergo morphogenesis and mineralization to become dorsal denticles (DD). (CD, Fand G to J) Their emergence precedes the subsequent eruption of adjacent parallel rows of the body dentition (BD). The dorsal denticles also begin to mineralize (H) before the development of the body dentition (I). The dorsal denticles are longer and wider than the body denticles (E, F and J). RD modeling suggests that the diffusion and interaction of an activator and initiator line inhibitor representing dorsal denticles (K) may explain the standardization of surrounding body denticles (I and M). (A) to (C) are computed tomography (CT), (D) to (F) are scanning electron microscopy (SEM) images, and (G) to (J) show alizarin samples stained in red. See Materials and Methods for RD modeling details. Scale bars, 250 μm (D), 200 μm (E), 100 μm (F), 10 mm (G) and 400 μm (H to J).

This study investigates the pattern of epithelial appendages in an early divergent lineage with respect to tetrapods using the small-dog shark (Scyliorhinus cannula). Using a combination of RD modeling and gene expression analysis, we investigated the mechanism and underlying molecular basis of shark dentition standardization. We then use experiments to inhibit the pathway of small molecule genes to reveal the functional conservation of these genes. Finally, we use RD modeling to demonstrate that our experimental results conform to a standardization system similar to Turing. Instead of following a random distribution (30), we find that the development of the shark's dentition is underpinned by a precise patterning mechanism that begins early in development. This conserved system may underlie the development of a wide range of epithelial appendages, thus facilitating the evolution of several functional characteristics observed in all vertebrates.


RD simulation and gene expression analysis suggest that a Turing-like system underlies standard shark body dentition

We first investigate the morphogenetic pattern of shark denticles. Two rows of dorsal dental placodes are visible in stage 32 of development [~80 days postfertilization (dpf)] (Fig. 1A) (42), preceding the appearance of the body denticles (Figure 1, C, D, and F). Compared to the body denticles, the dorsal denticles are larger and wider and do not have distinct ridges associated with the reduction of hydrodynamic trawling (Figure 1, D to F) (5). Simulation of an RD model was used to determine if the dorsal lines of the dentifrice can act as "starter" lines, triggering the standardization of the surrounding body denticles. The patterns were generated from a row of primers representing the dorsal denticles (Fig 1K), of which waves of activating and inhibitory morphogens irradiated according to predefined values ​​(Fig. 1L and table S1; see Materials and Methods for more details). Points formed in lines adjacent and parallel to the initiator line. Upon reaching a steady state, the starter points remained larger than the newly formed points (Fig. 1M), reflecting the scaling of the shark (Fig. 1, D to J). This model provides theoretical support for a Turing-type system that controls the dental standard in sharks.

To compare the standardization of shark denticles and chicks' feathers, we examined the expression of β-catenin (β-cat), an early regulator of chick epithelial placode signaling (Fig 2 and Fig S1) (43). The chicken embryo expresses a dorsolateral β-cat on embryonic day 6 (E6) (Fig. 2, C and D). This range becomes compartmentalized in placements of individual feathers in E7 (Fig. 2, G and H), which trigger the emergence of adjacent parallel rows of placoid (Fig 2, K and L) (18). The lateral line of the shark expresses β-cat in stage 31 (~ 70 dpf), shortly before the beginning of the dentition standardization (Fig. 2, A and B). A continuous range of expression was not observed in the shark; however, two dorsolateral rows of dental placodes appeared simultaneously in stage 32 (~ 80 dpf), expressing β-cat (Fig. 2, E and F). These lines emerged parallel to one of the lateral lines (Fig. 2, A to F). The placodes of the lower body dentition subsequently emerged in rows adjacent to the dorsal denticles later in stage 32 (~ 100 dpf) (Fig 2, I and J). The dorsal shark denticles may be acting as rows of initiators, triggering the emergence of neighboring units in a Turing mechanism similar to the plumage pattern. Having noted this similarity between shark epithelial appendage patterns and chicks, we next examined the gene expression underlying a putative pattern system similar to Turing's in the shark.

Figure 2 Preserved primer lines can trigger surrounding epithelial placodes in the shark and cub.

Full-mount ISH for β-cat was performed during the standardization of the epithelial appendix of shark denticles (AN, B, AND, F, Iand J) and chicks' feathers (W, D, G, H, Kand I). At E6, the chick exhibits a continuous range of β-cat expression (C and D), which then becomes compartmentalized in feather placodes (G and H). This line of initiators triggers the appearance of surrounding feather placodes, following a RD system (17). (A and B) At stage 31 (~ 70 dpf), shark dentition placodes are not visible, although the lateral line sensory system pattern is demarcated by β-cat. (E and F) In stage 32 (~ 80 dpf), two dorsolateral rows of dental placodes are visible. (I and J) Subsequently in stage 32 (~ 100 dpf), the surrounding rows of body dental placodes also express β-cat. The rows of the shark's dental duct may be triggering the emergence of the dentiger from the body following a Turing-like system comparable to the plumage pattern. LL, lateral line; PA, body placode; P, placode. 1000 μm (B, C, G, J and K), 500 μm (D, F and H) and 750 μm (L).

Using in situ hybridization (ISH), we sought to identify the potential activators and inhibitors that make up this Turing standardization system. A set of genes was selected based on their importance during the plumage pattern (18), and its expression was analyzed along the shark scaling (Fig 3 and Fig S1). In stage 31 (~ 70 dpf), no placodes of the dorsal dentition were detected (Figure S2), although β-cat marked expression development of the lateral line sensory system (Fig. 2, A and B). In the initial stage 32 (~ 80 dpf), two dorsolateral rows of dental placenta were visible, expressing the known plumage pattern activators, fgf4 and shh, as well as the inhibitor bmp4 (Fig. 3, A to C) (18, 42). Similar to the feather pattern, bmp4 was expressed within the placodes instead of the interplaced regions, suggesting that their inhibitory action is indirect (18). The mesenchymal marker of the development of feathery gems, fgf3, was also expressed in rows of dorsal denticles (Fig. 3D) (44), along with the transcription factor of the domain runx2 (Fig. 3E), which is associated with FGF signaling through morphogenesis and mineralization of other skeletal vertebrate elements (4547). A gradient anterior to posterior dorsal dentition development was observed.

Fig. 3 Conserved RD markers are expressed during standardization of the shark dentition.

The expression of genes thought to control the RD pattern of chicks' feathers was mapped during shark dentition standardization (17). (AN for W) In stage 32 (~ 80 dpf), the placodes of the dorsal dorsal shark fgf4 and shh, which are considered feather standardization activators, and bmp4, which is considered an inhibitor (17). (D and AND) Dorsal rows also expressed fgf3, a dermal marker of the development of feathered gemstones and runx2, which is associated with FGF signaling during the development of mammalian teeth (44, 45). (F for OLater, in stage 32 (~ 100 dpf), these genes are expressed during standardization of adjacent and parallel rows of placodes of the dentiger of the body. (P for R and TThe ISH section of the body denticles revealed epithelial shh and mesenchymal expression of fgf4, bmp4and runx2. (s) Expression of fgf3 was observed in the epithelium and mesenchyme. White dashed lines separate columnar cells from the basal epithelium and the underlying mesenchyme. Scale bars, 500 μm (A to E), 2000 μm (F to J), 1000 μm (K to O) and 50 μm (P to T).

Subsequently, in the developmental stage 32 (~ 100 dpf), the placentas of the body dentition become visible in the adjacent rows and parallel to the rows of the dorsal dentition. The body denticles extend throughout the ventral trunk and eventually propagate to the entire flank and ventral surface. We understand that there are several initiation sites (48), which are important for the extension of the dental pattern to the extremities, such as the paired flippers. The redistribution of the same set of genes expressed during the development of the dorsal dentition was observed during the standardization of these small denticles of the body (Fig. 3, F to O). ISH Section revealed that shh was expressed in the dentition epithelium of the body, while fgf4, bmp4and runx2 were expressed in the underlying mesenchyme (Fig. 3, P to R and T). The expression of fgf3 was observed in both the epithelium and the mesenchyme (Fig. 3S). Overall, these results revealed extensive conservation of RD-related gene expression between dental and plumage patterns (18, 43, 49).

The RD-related genes are functionally conserved during the standardization of the denticles of the shark body

To verify the functional conservation of the expressed genes during the dental standard, we performed inhibition experiments on the gene pathway of small molecules. The embryos were treated with beads loaded with the FGF receptor inhibitor SU5402 (50) or dimethylsulfoxide (DMSO) as the control. The beads were implanted below the epithelium in embryos of stage 31 (~ 75 dpf), adjacent to the rows of primordia of the emerging dorsal dentition (Fig. 4A). Development continued before the genetic and phenotypic effects of treatment were examined at various times.

Fig. 4 Tumor inhibition experiments reveal the functional conservation of RD associated genes.

(AN) Granules loaded with the FGFR inhibitor SU5402 were implanted under the epithelium of shark embryos at 75 dpf. (W for NFirst, we analyzed the gene expression at 5 dpt. We propose that the breakdown of a conserved system of activator-inhibitor feedback between fgf4, shhand bmp4 (B) led to the localized downward regulation of both shh and bmp4, resulting in atrophied growth of dorsal dorsal primordia, highlighted by black and white arrows (C to J) (17). (K to N) Expression of spry2, a transcriptional reading of FGF signaling was also reduced (50). We observed localized inhibition of gene expression at 5 dpt in all SU5402 beads (n = 5/5) and without DMSO control samples (n = 5/5). (O) Expression of fgf4 at 25 dpt showed that this inhibition resulted in a gap in the dorsal line of the dentition, which was occupied by smaller denticles of the body (n = 2/2). (P) No gap was observed in control samples of DMSO (n = 2/2). Red alizarin staining revealed that this range was maintained in 75% of SU5402 treated dorsal lines at 50 dpt (n = 6/8), whereas no gap was observed in rows treated with DMSO control cord (n = 7/7) (Figure S5). (Q) This pattern was maintained on the dorsal lines with SU5402 rim at 75 dpt, once the body denticles began to mineralize (n = 7/8). (R) Control samples of DMSO did not shown = 9/9). The output of the RD simulation including a gap in the primer line (s) was consistent with the observed experimental pattern; smaller units occupied the gap in the queue (T and you). The black dashed lines show the location of the vibrational sections of the ISH total assembly (E, F, I, J, M, and N). E, F, I, J, M and N), 300 μm (O and P) and 400 μm (Q and R) ).

First, the ISH for RD-related genes was performed 5 days after treatment (dpt). Localized inhibition of shh and bmp4 the expression was observed in dorsal dorsal placodes treated with SU5402 beads, whereas expression was unchanged in rows treated with DMSO beads (Figures 4, C through J and Figures S3 and S4, A through D). We propose that inhibition of FGF signaling disrupted a conserved system of activator-inhibitory feedback between fgf4, shhand bmp4, which similarly mediates the feather pattern (Fig. 4B) (18). In addition, we observed down-regulation of 2 (2)spry2) expression (Fig. 4, K to N). As spry2 is a transcriptional reading downstream of FGF signaling (51), this supports the idea that treatment with SU5402 led to inhibition of FGF in this system. Sections of full-mount ISH samples revealed a delay in the development of early dentition (Figure 4, C to N and Figure S4), suggesting that inhibition of FGF signaling during early morphogenesis is sufficient to restrict the growth of the dorsal dentition . As the dorsal denticles develop in an anterior to posterior gradient, the treatment effect was stronger in units that underwent early morphogenesis at the time of filling, rather than simply the units closest to the cord (Fig. N). For example, in Fig. 4K, the account is previously positioned to units with reduced gene expression, since the units closest to the account are more advanced in their development. Subsequent units submitted to early morphogenesis (marked with black arrowhead) were affected by the treatment. Embryo growth can also affect the proximity of the cord to the area of ​​inhibition. These results suggest that there is a functional conservation of a regulatory network of the central gene that controls the dental pattern of sharks, with FGF signaling playing an important activating role.

Next, we examined the effect of bead implants at 25 dpt, the stage at which minor body dentures start (~ 100 dpf). Using ISH, we visualize fgf4 expression to examine how the disruption of dorsal dentition development altered the subsequent pattern (Fig. 4, O and P, and Figs. 4, E and F). The dorsal dorsal primordia did not undergo morphogenesis after inhibition of FGF, resulting in a gap in the row. This gap was filled by small placodes of the body (Fig. 4O), potentially as an inhibitory field around the dorsal dentition that did not extend to this area. In contrast, control samples exhibited a complete line of dorsal denticles (Fig. 4P). The alizarin red staining of samples of SU5402 beads fixed at 50 and 75 dpt revealed that this pattern was maintained during development, with smaller and mineralized denticles occupying the slits in the rows of the duct (Figures 4, Q and R and figs. S5 and S6). Next, we examined whether this standardization response was consistent with a DR system. Therefore, we simulated the RD model (Fig. 1, K to M) with a unit missing from the primer line (Fig. 4S) to mimic the functional experiment. The model output showed remarkable similarity to the standard after cord implantation, with smaller units occupying the space resulting from the missing primer point (Fig. 4, T and U). These results provide further evidence that a Turing-type system controls shark dentition standardization, since the model response remains robust after experimental manipulation.

The re-fitting of the RD model may explain the diversity of dentition patterns

Having found evidence for the standardization of Turing-like dentition in the feline, we sought to examine the role of this system in other species of elasmobranchs. Among the elasmobranchs, tooth density is diverse, with most sharks showing relatively dense coverage. Comparatively, the thornback skateRaja clavata) and the duckling (Leucoraja erinacea) is increasingly sparse (Fig. 5, A to F). We returned the parameters of inhibitory morphogens and activators in the RD model to predict this diversity in dentin density of elasmobranchs.

Fig. 5 Changes in RD parameters may explain the diversity of dentition patterns.

(AN for F) The diversity of denticles varies between elasmobranchs, with standardization becoming denser in relation to the hawk (S. canicula) to skate thornback (R. clavata) and the duckling (L. erinacea). (G) Parameters of the RD model were initially defined to result in standardization similar to a cat. (H) Decreased rate of constitutive degradation of the inhibitor (dv) and maximum net production rate (Gmaximum) while increasing its diffusion coefficient (Dv) resulted in a less dense skate pattern. (E) The starting points have been raised and placed farther away to reflect the dorsal line of the skateboard. (I) Decrease in the constitutive production rate of the activator (Wyou) further reduced the coverage density, resulting in a bit of pattern similar to that of skateboarding. See Materials and Methods for RD modeling details and table S1 for specific parameter values. Scale bars, 400 μm (D) and 1000 μm (E).

The model parameters were initially adjusted to result in a feline-like dentition pattern (Figs 1, K to M and 5, D and G). The rate of constitutive degradation of the inhibitor (dv) and maximum net production rate (Gmaximum) were then decreased, while their diffusion coefficient (Dv) was increased (Table S1). Initiator spots were enlarged and spaced to reflect the dorsal line of the skate (Fig. 5E). This led to a decrease in coverage density, giving rise to patterns comparable to those of skateboarding spines (Fig. 5, E and H). Then, the activator's constitutive production rate (Wyou) was decreased (Table S1). This further reduced the coverage density, giving rise to patterns comparable to those of the small skate (Fig. 5, F to I). It is worth noting that numerous alternative combinations of parameter values ​​can result in outputs similar to those shown here (Fig. 5, G to I), as well as much more diversified outputs (9). Overall, these results demonstrate that simple changes in the RD model parameters can give rise to a wide diversity of pattern results comparable to those observed in existing elasmobranch species. The plasticity of this system may underlie broad variations that cover the vast spectrum of vertebrate epithelial appendage patterns.


Our results provide both theoretical and experimental evidence to suggest that standardization of the shark dentition is controlled by a system similar to the conserved Turing, also known to mediate the plumage pattern of the chicks (18). This mechanism probably controlled the development of the epithelial appendix for at least 450 million years, spanning the evolution of vertebrates from sharks to mammals (9, 21, 28). This system includes a dorso-lateral initiator line that triggers the emergence of adjacent appendages, controlled by functionally conserved activators and inhibitors, including fgf4, shhand bmp4 (18). In addition, we show that the alteration of the parameters of this system can explain the diversity of the dental pattern observed between different species of elasmobranchs.

Previous experimental work investigating RD standardization has focused extensively on its role throughout amniotes, specifically mice and chicks (18, 21). In addition, rearrangement of zebrafish pigmentation after partial-band ablation is concomitant with a DR system (52). The standardization of the dentition after cord implantation showed remarkable similarity with this experiment (Fig. 4); in both systems, the space in the original line was filled by adjacent lines. We provide evidence for the Turing-like pattern in chondrocytes. This supports both experimental and theoretical work, suggesting that Turing's standardization is of great importance throughout the evolutionary history of vertebrates and is common to groups of taxonomically diverse vertebrates (9).

In addition, we demonstrated that changes in the parameters of this system may explain the diversity of patterns of epithelial appendages between different species (Fig. 5). Within elasmobranchs, this may have facilitated the evolution of various species-specific dental functions, including protective shielding, hydrodynamic drag reduction, feeding and communication (5, 7, 29, 33, 53). More broadly, this system may underlie the patterns of epithelial appendages in all other vertebrates. For example, DR can control the mammalian capillary density, which is closely linked to thermoregulation (6). Small changes in this conserved system can support the diversity of vertebrate patterns.

Future research should address the formation of primer lines that trigger subsequent Turing patterns (Fig. 2). In the pinto, this line originates as a continuous band, which then forks in two lines, before the expression becomes localized in individual feather placodes (54). The shark has two rows of denticular placodes (18, 39), suggesting that the single line of initiation of bifurcation of the chick may be a derived characteristic. Sequencing of the transcriptome demonstrated that genes associated with neural development are significantly up-regulated in the skin during the chick primer line pattern. This is indicative of the synchronicity of development between the nervous system and the plumage pattern (55). The lateral line of the shark is a system of innervated sensory organs that appear parallel to subsequent placodes in the dorsal line (Fig 2, A and B and Fig S2D). It is possible that these systems are synchronous in the shark, with the lateral line mediating the standardization of the line of the shark's dental initiator. In addition, the lateral line extends the entire length of the body and can mediate the Turing pattern posterior to the dorsal rows, which extend approximately halfway along the dorsal trunk. In addition, there are several breeding sites for patterns, including those located on the wings and the pectoral fins of the puppy and elasmobranchs, respectively (48, 56). If these sites have individual primer lines it is unknown, presenting a gap in our understanding of pattern initiation.

The importance of DR-controlled standardization has been debated for a long time (9). However, there is a growing body of theoretical and experimental work supporting the relevance of this model (11, 12, 1821). Nossas descobertas fornecem suporte para esta pesquisa, demonstrando que um antigo sistema de Turing-like controla padrões de apêndices epiteliais em condroitiosanos, que pertencem a uma linhagem divergente precoce, com relação aos tetrápodes. Sugerimos que diversos grupos de vertebrados compartilham esse mecanismo de padronização comum e conservado, antes que o desvio na morfogênese posterior origine apêndices tegumentares específicos do clado, como dentículos, penas e cabelos.


Tubarão e pintinho

A Universidade de Sheffield é um estabelecimento licenciado sob a Lei de Animais (Procedimentos Científicos) de 1986. Todos os animais foram abatidos por métodos aprovados citados no Anexo 1 da Lei. Ovos de galinha castanha Bovan fertilizados (Henry Stewart & Co., Norfolk, RU) foram incubados a 37,5 ° C antes da fixação durante a noite na solução de Carnoy entre E6 e E9. S. canicula embriões (North Wales Biologicals, Bangor, Reino Unido) foram criados em água salgada artificial oxigenada (Instant Ocean) a 16 ° C. Os embriões de tubarão foram sacrificados com MS-222 (tricaina) a 300 mg / litro e fixados durante a noite em paraformaldeído a 4% em solução salina tamponada com fosfato (PBS). Após a fixação, os embriões de frango e tubarão foram desidratados através de uma série graduada de PBS para etanol (EtOH) e armazenados a -20 ° C.

Micro-CT e SEM

A digitalização de micro-CT de alta resolução foi realizada usando um scanner Xradia Micro-XCT no Imaging and Analysis Center, Museu de História Natural, Londres. S. canicula Os embriões foram corados com ácido fosfotúngstico a 0,1% em EtOH a 70% durante 3 dias para aumentar o contraste. As varreduras foram renderizadas usando a ferramenta de exploração de volume tridimensional Drishti ( O SEM foi realizado usando um scanner Hitachi TM3030Plus SEM de bancada a 15.000 V.

Alizarina vermelho claro e manchado

Embriões foram reidratados de EtOH para PBS e corados durante a noite com alizarina vermelha em hidróxido de potássio (KOH), como descrito anteriormente (4). As amostras foram visualizadas em glicerol utilizando um estereomicroscópio Nikon SMZ15000. Barras de escala foram criadas em Fiji (57).

Modelagem RD

A modelagem RD da modelagem dentária do corpo de tubarão foi realizada usando um modelo ativador-inibidor proposto por Kondo e Miura (9) baseado nas equaçõesImagem Incorporada(1)Imagem Incorporada(2)Onde you(t, x, y) e v(t, x, y) denotam as concentrações de um ativador e inibidor, respectivamente, no tempo t e localização (x, y). As equações 1 e 2 descrevem a taxa de mudança dessas concentrações no tempo e no espaço devido à difusão e produção regulada e degradação das espécies moleculares. As funções não lineares F(you, v) e G(you, v) são definidos porImagem Incorporada(3)Imagem Incorporada(4)

As equações 1 e 2 foram resolvidas no domínio quadrado bidimensional 0 < x < I0 y < I por tempos 0 < t < T sujeito a condições de contorno sem fluxo e condições iniciais prescritas que variavam entre as simulações. Para as simulações mostradas nas Figs. 1 (K a M) e 5 (G a I), a condição inicial foi dada porImagem Incorporada(5)Imagem Incorporada(6)onde cada um (xI, yI) defines the center of a spot in an initiator row representing dorsal denticles of a given number (nlocal) and radius (Rlocal). Figures 1 (K to M) and 5G were generated using Rlocal = 4.5, nlocal = 6, and (xi, yi) = (iL/5, L/2). Figure 4 (S to U) was generated using the same initial condition but with the spot centered at (x2, y2) removed. Figure 5H was generated using Rlocal = 5.25, nlocal = 3, and (xi, yi) = ((3i + 2)L/10, L/2), reflecting fewer, larger, more widely spaced initiator spots.

The RD model was solved numerically using an explicit finite difference method, choosing a spatial discretization Δx and sufficiently small time step Δt to ensure numerical stability. Python code to generate Figs. 1 (K to M), 4 (S to U), and 5 (G to I) is provided in the Supplementary Materials. The parameter values used to generate Figs. 1 (K to M) and 4 (S to U) were given by du = 0.03, Du = 0.02, Theu = 0.08, bu = − 0.08, cu = 0.04, Fmaximum = 0.2, dv = 0.08, Dv = 0.6, Thev = 0.16, bv = 0, cv = − 0.05, and Gmaximum = 0.5, with a domain of size L = 75, end time T = 1500, spot radius R = 4.5, initial concentration u0 = 5, and discretization Δx = L/128 ≈ 0.58, Δt = (Δx)2/8Dv ≈ 0.07. These values were chosen on the basis of an ad hoc exploration of parameter space around parameter values previously identified by Kondo and Miura as leading to patterning (9). Parameter values for Fig. 5 are given in table S1. For Fig. 5 (H and I), because the value of Dv was reduced, we updated the value of Δt = (Δx)2/8Dv ≈ 0.04 to maintain numerical stability.

In situ hybridization

Digoxigenin-labeled antisense riboprobes were designed using partial skate (L. erinacea) and catshark (S. canicula) EST (expressed sequence tag) assemblies (SkateBase; (58), the Vertebrate TimeCapsule (VTcap;, and transcriptome data from RNA sequencing (unpublished). Sequences of forward and reverse primers (Sigma) are as follows: chick β-cat, TCTCACATCACCGTGAAGGC (forward) and CCTGATGTCTGCTGGTGAGG (reverse); shark β-cat, GGTGAAAATGCTTGGGTCT (forward) and GGACAAGGGTTCCTAGAAGA (reverse); Tubarão fgf4, ATGTTGATCAGGAAGCTGCG (forward) and GTATGCGTTGGATTCGTAGGC (reverse); Tubarão shh, TGACTCCCAATTACAACCCGG (forward) and TCAGGTCCTTCACTGACTTGC (reverse); Tubarão bmp4, GATCTCTACAGGCTGCAGTCC (forward) and GATCTCTACAGGCTGCAGTCC (reverse); Tubarão fgf3, CTTGCTCAACAGTCTTAAGTTATGG (forward) and CGGAGGAGGCTCTACTGTG (reverse); Tubarão runx2, ATCTCTCAATCCTGCACCAGC (forward) and CCAGACAGACTCATCAATCCTCC (reverse); and shark spry2, AACTAGCACTGTGAGTAGCGG (forward) and GTTCCGAGGAGGTAAACTGGG (reverse). Riboprobes were synthesized using the Riboprobe System SP6/T7 Kit (Promega) and DIG RNA Labeling Mix (Roche). Whole-mount and section ISH was performed as previously described (4, 59). To compare sequences between the chick and shark, phylogenetic gene trees were reconstructed from protein coding sequences extracted from, aligned to S. canicula sequences obtained during probe synthesis (see fig. S1 for details) (60, 61). Whole-mount ISH samples were imaged using a Nikon SMZ15000 stereomicroscope, and sections were imaged using an Olympus BX51 microscope and Olympus DP71 Universal digital camera attachment. Vibratome sections shown in Fig. 4 were cut at a thickness of 30 μm. Adjustments to image contrast and brightness were made to improve clarity. Scale bars were added using Fiji (57).

Bead implantation experiments

Embryos were treated with Affi-Gel Blue beads (Bio-Rad) loaded with SU5402 (2 mg/ml; Sigma) in DMSO. Control beads were loaded with DMSO. Stage 31 (~75 dpf) embryos were removed from their egg cases and anaesthetized before beads were surgically implanted using sharpened tungsten wire. Embryos were then cultured in six-well plates with artificial salt water and 1% penicillin-streptomycin (Thermo Fisher Scientific). At stage 32 (~100 dpf), embryos were transferred to 70-ml plastic containers (Sarstedt) floating in a 200-liter tank. The number of replicates and observed effects for different analyses are shown in Table 1.

Table 1 Summary of the number of replicates for bead inhibition experiments (shown in Fig. 4 and figs. S4 and S5).


Supplementary material for this article is available at

Fig. S1. Phylogenetic gene trees reconstructed from protein coding sequences extracted from

Fig. S2. Dorsal denticle placodes are not visible at stage 31 (~70 dpf).

Fig. S3. Individual vibratome section images comprising false-colored ISH composite images.

Fig. S4. Replicates of beaded shark embryos after whole-mount ISH.

Fig. S5. Replicates of clear and stained shark embryos showing RD response to SU5402 beading.

Fig. S6. SEM images of shark embryo 75 days after beading.

Table S1. Activator and inhibitor values for RD model.

Python script for RD simulations

This is an open-access article distributed under the terms of the Creative Commons Attribution license, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Acknowledgments: We would like to thank N. Monk (School of Mathematics and Statistics, University of Sheffield) for initial discussions regarding RD modeling. We also extend our gratitude to K. Martin, Z. Johanson (Department of Earth Sciences, Natural History Museum, London), F. Ahmed, and A. Garbout (Imaging and Analysis Centre, Natural History Museum, London) for assistance with micro-CT imaging. Thornback ray (R. clavata) embryos were donated by the Native Marine Centre, Weymouth, UK, and little skate (L. erinacea) embryos were sourced from the Marine Biological Laboratories (MBL), Woods Hole, MA, USA. Last, we thank M. Placzek for the gift of Affi-Gel Blue beads (Bio-Rad) and K. Onimaru for advice on experimental methods. Funding: This research was supported by the following research grants: Natural Environment Research Council (NERC) Standard Grant NE/K014595/1 (to G.J.F.), NERC PhD studentship (to L.J.R.), and Leverhulme Trust Research Grant RPG-211 (to G.J.F.). This work was also funded through “Adapting to the Challenges of a Changing Environment” (ACCE), a NERC-funded doctoral training partnership (to R.L.C.) ACCE DTP (NE/L002450/1). A.G.F. is supported by a Vice-Chancellor’s Fellowship from the University of Sheffield. Author contributions: R.L.C. and G.J.F. designed the project. D.J.D. and R.L.C. undertook initial RD modeling. A.G.F. refined the RD modeling and wrote the Python RD simulator. L.J.R. undertook section ISH (shown in Fig. 2). A.P.T. created gene trees shown in fig. S1. R.L.C. performed all other experimental work and data collection. R.L.C. and G.J.F. analyzed and interpreted the results. R.L.C. and G.J.F. wrote the manuscript. All authors read, edited, and approved the final manuscript. Competing interests: The authors declare that they have no competing interests. Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials. Additional data related to this paper may be requested from the authors.


Source link