Stickleback pharyngeal tooth morphogenesis. All images show sagittal or transverse sections from sticklebacks (20–40 mm standard length). The basalmost layer of epithelium is outlined with yellow dashed lines in all panels, the mesenchyme is outlined in red. During late bell stages (e, k), the epithelium towards the apex of the forming tooth becomes thickened, no longer presents a clear bilayer, and will eventually be punctured during tooth eruption (gray outlines). a–f Staging series of hematoxylin and eosin-stained transverse sections through stickleback pharyngeal tooth germs at the stages indicated. Epithelium (“epi”) and mesenchyme (“mes”) are labeled in a. g–l Staging series of DAPI-stained sagittal sections through stickleback pharyngeal tooth germs at the stages indicated. Gray portions of dotted epithelium lines in J indicate the epithelium/tooth connection that was present on an adjacent section. Bone matrix is false-colored red in k and l. Scale bars = 20 μm

A comparison of zebrafish and stickleback pharyngeal tooth replacement. a, b Zebrafish and stickleback first tooth replacement events, respectively. Alizarin stained and dissected ceratobranchial 5 elements and their teeth from both species at 4 dpf zebrafish (a) and 30 dpf stickleback (b). Anterior to top. Arrows indicate first replacement tooth germs (bilaterally paired). Note that the zebrafish germs form on the ventral/medial side of the predecessor pioneer tooth (which in this case is the middle of 3 the ankylosed teeth on each side, tooth 4v1). Inset box in b magnified in b’ shows first replacement tooth. c, d Replacement histology on the coronal (d) and oblique coronal (c) axes (see Methods). Arrowheads mark the putative successional dental epithelium (SDE) in each image. In zebrafish, this tissue takes the form of a true successional dental lamina (SDL). In sticklebacks, there is only a subtle, non-elongated epithelial thickening that surrounds the tooth shaft like a collar, immediately beneath the rest of the nearby epithelial tissues (see Additional file 1: Fig. S1 for sagittal view). Arrows mark tooth germs at the stages indicated. A bracket marks the tight epithelial association between an early tooth germ and the position of the presumed SDE. Scale bars in a and b = 100 μm; c and d = 20 μm

Oral and pharyngeal tooth replacement histology in stickleback. In all panels, dotted circles mark the general location of erupted tooth dislodgement adjacent to a presumed replacement tooth germ, arrows mark tooth germs, and carets indicate likely osteoclasts. All panels show sagittal (a, b) or transverse (c, d) sections and are oriented with the dorsal side facing upwards. a, b Oral tooth fields on the premaxilla and dentary bones were only observed as replacing via a 1-for-1 mechanism. c Pharyngeal replacement tooth germs were typically observed as dislodging only one other erupted pharyngeal tooth in 74.6% of cases (n = 50/67, see Methods). d By contrast, some pharyngeal replacement tooth germs appeared to be dislodging two erupted teeth simultaneously in 25.4% of cases (n = 17/67, see Methods). All scale bars = 20 μm

Wnt signaling in zebrafish and stickleback replacement tooth germs. Black arrows mark replacement tooth germ epithelium, white arrows mark replacement tooth germ mesenchyme. awnt10a expression in a cap-stage zebrafish replacement tooth germ. bWnt10a expression in a bud-stage stickleback replacement tooth germ. Note concentrated expression in the distalmost epithelium (arrow). clef1 expression in a cap-stage zebrafish replacement tooth germ. dLef1 expression in a bud-stage stickleback replacement tooth germ. e, f TCF/Lef synthetic Wnt activity reporter is active in adult replacement tooth germs. Scale bars in a–d = 20 μm; e and f = 500 μm

Successional dental epithelium (SDE) expression in zebrafish and stickleback. In all panels, a dotted line demarcates the basalmost pharyngeal epithelial layer, which is made gray when missing from a particular section (e.g., when it is interrupted by an erupted tooth). a Expression of bmp6, bmpr1aa, cd34, gli1, igfbp5a, nfatc1, lgr4, lgr6, pitx2 was assayed by in situ hybridization (ISH) in zebrafish. Expression of all 9 genes was detected in the successional dental lamina (SDL—black arrowheads). b Expression of Bmp6, Bmpr1a, CD34, Gli1, Igfbp5a, Lgr4, Lgr6, Nfatc1b, and Pitx2 was assayed by ISH in stickleback. Expression of all nine genes was detected in the proposed “successional dental epithelium” (SDE) of stickleback. Black arrowheads mark the SDE. Bmpr1a and Gli1 were additionally expressed in the superficial oral epithelium surrounding erupted teeth (white arrowheads). Tooth germs are indicated with a black arrow. All scale bars = 20 μm

Schematic of successional dental epithelia in different tooth regeneration systems. Each of the four columns of illustrations represents an idealized transverse cross-section of a tooth position from each of the polyphydont vertebrate groups indicated at the top. Each row of illustrations shows a different phase of the regeneration process as indicated on the left. Mineralized tissues are shown in red, and epithelia are shown in yellow. The continuous basalmost layer of epithelium that abuts the basement membrane is additionally delineated with a dotted line. Arrows indicate tooth germs, arrowheads indicate the successional dental epithelium (SDE) in each species as suggested by our model. The epithelia are depicted slightly off-plane (shallower on the z-plane) with respect to the erupted tooth (analogous to viewing a thick section). The left two columns (stickleback and salmonids) illustrate species that lack a successional dental lamina (SDL), whereas the right two columns illustrate species that do have an SDL; the zebrafish, which exhibits a transient SDL (note lack of SDL in the bottom row) vs. salamanders, which maintain a permanent SDL (arrowhead in bottom row). The stickleback drawings are representative of both oral and pharyngeal tooth regeneration and based on our own observations. The zebrafish pharyngeal tooth drawings are based on our own observations in conjunction with previous literature [42, 69]. The salmonid drawings are based on data from rainbow trout [47, 49] and Atlantic salmon [48, 136], and are based primarily on oral tooth histology (though pharyngeal teeth are described as behaving similarly in rainbow trout [49]). The salamander oral tooth drawings are based on data from the Iberian ribbed newt [137, 138] and axolotl [139].