Sage Journals: Discover world-class research

Abstract

The odontoclast is a rarely studied cell type that is overly active in many dental pathologies, leading to tooth loss. It is difficult to find diphyodont mammals in which either physiological or pathological root resorption can be studied. Here we use the adult leopard gecko, which has repeated cycles of physiological tooth resorption and shedding. RNA-seq was carried out to compare gene expression profiles of functional teeth to developing teeth. Genes more highly expressed in bell-stage developing teeth were related to morphogenesis (PTHLH, SFRP2, SHH, EDAR). Some genes expressed in osteoclasts (ACP5, CTSK, CSF1R) were relatively more abundant in functional teeth compared with developing teeth. There was, however, no differential expression of RANKL (TNFSF11) in the 2 tooth types. In addition, functional teeth expressed proteolysis genes not found in osteoclasts (ADAMTS2, 3, 4, 14; CTSA, CTSH, CTSS). We used tartrate acid resistant phosphatase and cathepsin K (CTSK) staining to identify odontoclasts in and around the gecko dentition. There were 3 populations of CTSK cells: (1) large, functional multinucleated odontoclasts in the crown of the tooth with a ruffled border inside resorption pits; (2) smaller, precursor cells in the pulp with fewer nuclei; and (3) flattened external precursor cells next to the root and bone of attachment. We found a positive relationship between developing teeth and the population of CTSK+ cells on the root surface. We tested a candidate signal that may be involved in CTSK+ cell presence. An antagonist of CSF1R was delivered to developing teeth in vivo, which resulted in a significant decrease in CTSK and CSF1R compared with DMSO controls. Thus, the CSF1 signaling pathway is upstream of CTSK in teeth. This is the first work to detail the molecular characteristics of odontoclasts during physiological tooth shedding and to demonstrate that in vivo, local drug delivery is possible in the gecko model.

Keywords

root resorption polyphyodonty reptile gecko RNA-seq cathepsin K

Introduction

Tooth replacement occurs once in humans and carnivores (diphyodont), but most vertebrates have tooth replacement throughout life (polyphyodonty) (Henriquez and Richman 2024). Instead of examining tooth initiation and development, here we focus on a less studied part of the tooth life cycle: the resorption of roots followed by shedding. Humans shed their teeth on a schedule, and odontoclasts are responsible for physiological root resorption (Domon et al. 1994; Sahara et al. 1996; Takada et al. 2004). Unwanted odontoclast activity occurs after orthodontic treatment, trauma, and pulpal inflammation, which increases the risk of root resorption and ultimately tooth loss (Iglesias-Linares and Hartsfield 2017). The prevailing assumption is that odontoclasts are just a different version of osteoclasts but with a different target tissue (Xiao et al. 2022). However, others have shown that some aspects of odontoclast biology are not the same as osteoclasts (Takada et al. 2004; Ogawa et al. 2019). To determine whether odontoclasts are indeed similar to osteoclasts, it is necessary to study physiological root resorption.

Here we developed an unconventional reptilian model that has polyphyodonty in order to study odontoclasts. As described for more than 100 y (Woerdeman 1921), reptilian teeth are resorbed and replaced in a periodic pattern independent of wear or injury. Four reptiles have been used to study molecular aspects of tooth development but not tooth resorption in adults: leopard geckos (Eublepharis macularius; Handrigan et al. 2010; Brink et al. 2021), bearded dragons (Pogona vitticeps; Salomies et al. 2019), iguanas (Iguana iguana; Brink et al. 2020), and alligators (Alligator mississippiensis; Wu et al. 2013). These studies confirm that signaling molecules in teeth are highly conserved in comparison with the mouse. Furthermore, when agonists or antagonists are applied to the dentition of reptiles, the expected response genes are activated or inhibited (Handrigan and Richman 2010; Wu et al. 2013). In contrast to these other reptiles, the leopard gecko has many teeth per quadrant that are exchanged rapidly (teeth per quadrant: 40 in geckos, 21 in alligators, 19 to 20 in iguanas; replacement interval: 4 to 5 wk in geckos, 4 to 6 mo in iguanas, and 10 to 20 mo in alligators). The functional teeth in geckos have a pleurodont attachment where the labial and lingual side of the root dentin merges with bone (bone of attachment). The lingual side of the root is much longer than the labial side is. In addition, the lingual side of the root is connected to the dental lamina, an epithelial layer connecting all teeth throughout the jaw. In contrast to mammals and alligators (Bertin et al. 2018), the developing teeth of geckos are not embedded in bony sockets but instead are attached to soft tissues and can be easily removed or treated (Brink et al. 2021). Since geckos shed their teeth in regular waves, there are odontoclasts present throughout the dentition similar to lizards and snakes (Fuenzalida et al. 1999; LeBlanc et al. 2023).

Here we carry out a transcriptome analysis of functional and developing teeth extracted from the jaw. Previous reptilian transcriptomes profiled only developing teeth and did not include odontoclasts (Tsai et al. 2016; Salomies et al. 2019). We identified many proteolytic enzymes and other markers that were unique to gecko functional teeth, possibly representing the odontoclasts. Using cathepsin K (CTSK) antibodies, we identified 3 stages in the odontoclast life cycle from precursors in the connective tissue to entering the pulp of the functional tooth to resorption of dentin in the crown and root. Finally, we performed in vivo experiments that showed signals from the developing tooth regulated CTSK+ cells near the functional tooth.

Methods

Leopard gecko eggs (Just Geckos) were hatched and housed at the University of British Columbia as described (Brink et al. 2021) (ethics protocol No. A19-0308; A23-0292). This study complied with the ARRIVE guidelines 2.0. Teeth were removed under sedation as described (Appendix methods; Fig. 1A) and used for RNA-seq analysis (accession: GSE220653; aligned to the genome for Eublepharis macularius; GCF_028583425.1). The role of colony stimulating factor 1 (CSF1) in gene expression was studied by exposing the developing teeth to the CSF1R antagonist BLZ945 (20 mg/mL) or DMSO in vivo. The developing teeth were removed 24 or 72 h later for quantitative reverse transcription polymerase chain reaction (qRT-PCR) or for histological analysis. See the appendix for full experimental details.

Figure 1.

Differential expression analysis of gecko functional and developing teeth. (A) Schematic of the removal of adult gecko maxillary functional and developing teeth for RNA sequencing. Multinucleated, clastic cells are in the pulp of functional teeth. (B, B′) Hematoxylin and eosin staining of extracted developing teeth. (C, C′) Proliferating cell nuclear antigen and Hoechst staining of near-adjacent sections. (D, D′) Pan-keratin staining shows epithelium surrounding, and the apical end of the crown is included. (E) Hierarchical cluster of functional and developing teeth samples based on Spearman correlation. Animals used in RNA-seq experiments include LGA, B, C. (F) Principal component analysis of samples. (G) Volcano plot of differential expression comparing developing to functional teeth determined by the hisat-Deseq2 pipeline. Red spots and negative values indicate genes more highly expressed in functional teeth; blue spots and positive values indicate genes more highly expressed in developing teeth. (H) Validation with quantitative reverse transcription polymerase chain reaction of genes differentially expressed in RNA-seq data. Unpaired t test. Error bars = mean ±95% confidence interval. ab, aboral; cl, cervical loops; la, labial; li, lingual; mes, mesenchyme; o, oral; pu, pulpal. Scale bar in B, C, and D = 500 µm. Scale bar for B′, C′, and D′ = 100 µm.

Results

Differential Gene Expression in Developing versus Functional Gecko Teeth

We carried out bulk RNA-seq to compare the expression of genes in developing versus functional teeth. Functional teeth were separated from bone; therefore, no osteoclasts or osteoblasts were included in the dissection (Fig. 1A). The developing teeth were removed along with attached dental epithelium and mesenchyme (Fig. 1B, B′–D, D′), while excluding the successional and dental lamina. Quality control analysis of the expression data confirmed tooth types clustered together (Fig. 1E, F). In the PCA analysis, 2 samples (FT-4 and DT-4) clustered separately from the other samples along PC1 due to a batch effect but grouped with similar tooth types in PC2 (Fig. 1F). The analysis of 4 replicates had sufficient power to identify more than 3,000 genes. There were 1,745 genes more highly expressed in functional teeth and 1,355 genes more highly expressed in developing teeth (adjusted P < 0.05; Fig. 1G; Appendix Tables 5, 6).

Developing teeth expressed genes associated with tooth morphogenesis, differentiation, and proliferation (Appendix Fig. 1A; Appendix Table 5). Bell stage enamel organ genes included EDAR (Ectodysplasin A receptor), SHH (Sonic Hedgehog), PTHLH (Parathyroid Hormone Like Hormone; Appendix Fig. 2), and dental mesenchyme genes such as LHX8 (LIM homeobox 8) (Fig. 1G, H). Amelogenesis genes included AMBN (Ameloblastin), ENAM (Enamelin), AMELX (Amelogenin), AMTN (Amelotin), and MMP20 (codes for Enamelysin) (Fig. 1H). The developing teeth had higher expression of chemotaxis genes SEMA3F, 3G, 6A, and SLIT1 (Research data 1).

The most highly represented Gene Ontology (GO) function in developing teeth was cell division (Appendix Fig. 1A). The noncanonical, planar cell polarity WNT pathway was enriched in developing teeth including 2 WNT antagonists SFRP1 and SFRP2 (Fig. 1G, H). Several FGF ligands were also in the top 50 most highly expressed in developing teeth (FGF3, 10, 18), suggesting activity of this signaling pathway (Appendix Table 5).

Spatially Restricted Localization of SFRP2, PTHLH, and SEMA3E Transcripts in Developing Teeth

We mapped the expression of several genes with unknown expression patterns in gecko teeth using in situ hybridization. SFRP2 expression was abundant in the mesenchyme surrounding the cap-stage enamel organs and successional laminae (Fig. 2A–D). PTHLH was mapped to the stellate reticulum of developing tooth enamel organs (Fig. 2E, E′), which was in agreement with mammalian data (Philbrick et al. 1998). We found high levels of SEMA3E in the odontoblasts of functional and developing teeth (Fig. 2F, G) which would explain why we did not find differential expression using qRT-PCR (Appendix Fig. 2). In addition, there was strong expression of SEMA3E outside the teeth next to the root dentin and bone of attachment (Fig. 2F, G).

Figure 2.

SFRP2, PTHLH, and SEMA3E expression in gecko teeth. Frontal sections of gecko jaws hybridized to RNAscope probes where the signal is chromogenically detected (pink) and sections are counterstained with hematoxylin. (A) SFRP2 expression around a tooth cap stage. (B) SFRP2 expression around a bell-stage tooth bud. (C) SFRP2 expression around the tip of the successional lamina. (D) SFRP2 surrounding the successional lamina. (E, E′) PTHLH expression in the stellate reticulum of a late bell-stage tooth. (F) SEMA3E expression in a sagittal section of a functional tooth with strong signal in the polarized odontoblasts synthesizing dentin. (F′) expression of SEMA3E in the interdental connective tissue. (G) A similar pattern is seen in a developing tooth with a strong signal in functional odontoblasts near the dentinoenamel junction. Expression in the connective tissue next to the tooth (arrows). Scale bars in A, B, C, D, E′ = 20 µm. Bar in E = 100 µm. Bars in F, G = 50 µm. ab, aboral; d, dentin; e, enamel space; li, lingual; la, labial; o, oral; od, odontoblast; sl, successional lamina.

Comparison of Gene Expression in Odontoclasts in the Gecko Dentition to Osteoclasts

The GO Biological process enrichment in functional teeth included dentinogenesis genes SPARC (Secreted acidic cysteine rich glycoprotein or Osteonectin) and DMP1 (Dentin matrix acidic phosphoprotein 1) (Fig. 1G, H; Research data 1). Osteoclast induction genes CSF1 and CSF1R receptor and TNFRSF11A (codes for RANK receptor) were more highly expressed in functional teeth. In contrast, the soluble receptor TNFRSF11B (codes for Osteoprotegerin [OPG]) and the ligand TNFSF11 (RANKL) were not differentially expressed. We found expression of TNFSF11 in odontoblasts in functional and developing teeth as well as, dental mesenchyme, and enamel organs of developing teeth (Appendix Fig. 3A–E). Interestingly, there was no expression of RANKL in odontoclasts (Appendix Fig. 3C, C′, D). Functional teeth differentially expressed higher levels of genes involved in macrophage chemotaxis, replacement ossification (occurs after dental trauma), regulation of ruffled border, and osteoclast differentiation (Appendix Fig. 1B). A major gene required for osteoclast fusion, DCSTAMP, was expressed in functional teeth. The main transcription factor involved in differentiation of osteoclasts was also detected (NFATC1). Genes required for dissolution of bone matrix (CTSK, ACP5, MMP2, MMP9) were highly expressed in functional teeth (Table 1; Fig. 1G, H; Research data 1). ACP5 codes for tartrate acid resistant phosphatase (TRAP), which resorbs mineral. CTSK resorbs collagen after dissolution of mineral (Drake et al. 1996). MMP9 and CTSK are expressed in odontoclasts inside dentin resorption pits (Tsuchiya et al. 2008). The data support the idea that extracted functional teeth contain odontoclasts in addition to normal pulpal cells.

Table 1.

Genes More Highly Expressed in Functional Teeth from RNA-seq Analysis.

Category	Gene Symbol	Fold Change	P Adjusted	Function	PMID Evidence That Gene Is Expressed in Osteoclasts^a or Odontoclasts^b
Monocyte marker	CD68	−2.13	0.000	Expressed in monocytes and macrophages	2266187^a
Induction of osteoclasts from progenitor cells	TNFSF11 ^c (RANKL)	0.19	0.822	Osteoclasts are dependent on RANKL for differentiation	33144645,^b 12887401^a
	TNFRSF11A ^c (RANK)	−2.36	0.019	Osteoclast receptor induces differentiation	33144645^a
	TNFRSF11B (OPG)	−3.64	0.070	Decoy receptor for RANKL	33633362^b
	CSF1	−1.14	0.011	Survival, proliferative and predifferentiation cue	30971820,^a 2191302^b
	CSF1R ^a	−1.25	0.010	Receptor for CSF1	30971820^a
	OSTM1 ^c	−0.39	0.040	Osteoclast maturation and function	15108279^a
	TYROBP (DAP12)	−1.72	0.002	Osteoclast differentiation	33144645^b
Osteoclast fusion	DCSTAMP	−1.72	0.008	Fusion of osteoclast precursors	27018136^a
	OCSTAMP	−0.27	0.783	Multinucleated osteoclast formation	29533736^a
Osteoclast differentiation	NFATC1	−1.38	0.003	Terminal osteoclast differentiation	12479813,^a 33144645^b
	MAFB	0.99	0.002	Negative regulation of osteoclastogenesis	17158225^a
	CITED2	−1.41	0.000	Osteoclast cell fate commitment	33288951^a
	JDP2	−0.84	0.000	RANKL-mediated osteoclast formation	12707301^a
	SPI1 (PU.1)	−1.43	0.001	Required for early stages of osteoclastogenesis	33144645^b
	FOS	−1.62	0.059	required for NFATC1 induction	33144645²
	JUN	−0.63	0.032	Required for RANKL signal transduction	15314684^a
	JUNB	−0.82	0.000	Osteoclast differentiation and proliferation	14769860^a
	CEBPA	−2.34	0.000	Osteoclast commitment and differentiation	27129246^a
Osteoclast matrix degradation and acidification	ACP5 (TRAP)	−2.32	0.001	Dephosphorylation of osteopontin and bone sialoprotein b	33144645^b
Osteoclast matrix degradation and acidification	CTSK	−1.64	0.009	Degradation of collagen	33144645,^b 11870419,^b 37563130^a
	MMP9	−1.96	0.008	Extracellular matrix degradation	33144645,^b 12398942^b
	ATP6V0D2	−2.27	0.018	Extracellular acidification	17128270^a
	CA2 ^c	−1.00	0.000	Acidification	9665810^a
	TCIRG1 ^c	−0.85	0.017	Subunit proton pump	10888887^a
	CLCN7 ^c	−0.36	0.247	Endosomal/lysosomal Cl− channel; acidification of resorption lacunae	11207362^a
Osteoclast attachment	SPP1 (OPN)	−3.77	0.000	Binds to hydroxyapatite	12070270^b
	ITGAV, integrin αV	0.79	0.002	Binds to osteopontin	8952617^b
Others	CALCR	−5.45	0.009	Calcitonin ligand directly inhibits osteoclast function	33144645^b

Expressed or functions in osteoclasts.

Expressed in odontoclasts.

Causes autosomal-recessive osteopetrosis.

We encountered other genes responsible for proteolysis in the pulp of functional teeth (fibrinolysis, collagen catabolic process) (Appendix Fig. 1B; Appendix Table 7; research data file 1). Enzymes included MMP23B (FC = −2.47), ADAMTS2 (FC = −1.85), ADAMTS3 (FC = −1.76), ADAMTS4 (FC = −2.03), ADAMTS14 (FC = −2.37), CTSA (FC = −1.2), CTSH (FC = −1.2) (Wang et al. 2023), and CTSS (FC = −2.26) (Smyth et al. 2022). Expression of these genes or proteins has not been reported in osteoclasts (Drake et al. 1996; Tsukasaki et al. 2020), and their spatial localization in the gecko dentition needs to be verified.

We next used TRAP and CTSK staining to identify putative odontoclasts. The CTSK antibody recognizes osteoclasts in regenerating tails of anoles (Vonk et al. 2023), and we confirmed this antibody identified gecko odontoclasts (Fig. 3A–F). Prior to resorption, functional teeth have an intact odontoblast layer and bone of attachment merging root dentin with the bone (Appendix Fig. 4A, A″). Odontoclasts are positioned on the root dentin and bone of attachment of the apical third and dentin in the middle third (Appendix Fig. 5A, C,C″; Appendix Fig. 6C; Appendix Table 8). Then odontoclasts entered functional teeth through the apical end (Appendix Fig. 4A′), while no CTSK+ cells reached the coronal pulp (Appendix Fig. 4A″, B″′). Inside the pulp, there were 2 populations of CTSK+ and TRAP+ cells, smaller cells with fewer nuclei, and larger, multinucleated cells inside resorption pits (Fig. 3B, B′, C, C′, G–I; Appendix Fig. 4B, B″; App. Fig. 5A–C; Appendix Fig. 6A–C; Appendix Table 8). The smaller pulpal cells were the likely precursors of active odontoclasts (Appendix Fig. 5B′). Eventually, the full odontoblast layer next to the dentin was replaced by multinucleated, CTSK+, TRAP+ cells. The cells in resorption pits had ruffled borders, and the Golgi apparatus was located between the nucleus and ruffled edge (Appendix Fig. 5B′). The ruffled border indicated that the collagenous matrix was actively being resorbed.

Figure 3.

Histology and morphometry of clastic cells in the adult gecko dentition. Frontal sections of maxillary gecko teeth. (A–D′) Serial sections. (E, F) Sections from a different tooth family. (A, A′) The odontoblasts at the periphery of the pulp are replaced by odontoclasts that are in resorption pits (arrows). (B, B′) Odontoclasts express tartrate acid resistant phosphatase (TRAP), as do cells in the center of the pulp. (C, C′) Odontoclasts are positive for cathepsin K (CTSK) and are multinucleated in a split-channel view (white arrows, D, D′). (E, E′) TRAP+ cells that are flat are located apical to the functional teeth, between the developing tooth and bone. (F) Adjacent section to E showing that the flattened cells are CTSK+ (white arrows). (G) Average area of CTSK+ cells in 3 locations show the largest cells are in the resorption pits. (H) Higher values of aspect ratio indicate the cell is elongated. (I) Log₁₀ of the number of nuclei per square micron in CTSK+ cells showed the cells inside resorption pits had significantly greater numbers of nuclei. P values determined with 1-way analysis of variance and Tukey’s post hoc test. Error bars = mean ± 95% confidence interval. Points on graph represent means of 3 teeth for 3 animals. Each tooth had a minimum of 3 cells analyzed in each of the three locations. See Appendix Table 8 for details. ab, aboral; boa, bone of attachment; d, dentin; la, labial; li, lingual; o, oral; oc, odontoclast. Scale bars in A, B = 200 µm. Scale bars in A′, B′ = 50 µm. Scale bars in C, D, E = 100 µm. Scale bars in C′, D′, F = 20 µm. Scale bar in E′ is 10 µm.

The TRAP and CTSK+ cells attached to root dentin, bone of attachment, or bone had no visible ruffled border and had Golgi apparatus superimposed on the nuclei, which correlated with a lack of resorptive activity (Appendix Fig. 5C′, D). Transient CTSK expression was also found in a restricted group of secretory ameloblasts in the cusp tip (Appendix Fig. 7A–C′) and in the odontoblasts just at the onset of dentin secretion (Appendix Fig. 7D–F).

Selective Removal of Developing Teeth Reduced the Surface Covered with CTSK+ Cells

Next, we tested whether developing teeth provide signals that support the presence of CSTK+ cells. Our previous analysis showed that selective tooth removal delays de novo tooth formation until such time as the dental lamina heals (between 3 and 4 wk) (Brink et al. 2021). We divided the lingual root surface of functional teeth into thirds and measured the percentage of the surface that was covered CTSK+ cells (Fig. 4A). The oral third encompasses dentin close to the crown of the tooth, the middle third includes the root dentin, and the aboral third consists of apical root dentin and bone of attachment. The distance covered by CTSK+ cells was not significantly different between conditions when considered whole (Fig. 4B). However, the middle third was covered with significantly more CTSK+ surface than the oral third was (Fig. 4C; Appendix Fig. 8A, B; Appendix Table 9). Selective removal of developing teeth resulted in significantly less of the middle and aboral third surfaces covered by CTSK+ cells (Fig. 4C, P = 0.0359; Appendix Fig. 8C, D; Appendix Table 9). Therefore, removal of developing teeth either prevented survival or reduced migration or the generation of new odontoclasts attached to the root dentin.

Figure 4.

Effect of tooth removal and in vivo drug treatment on CTSK+ cells. (A) Measurement scheme for the distance along the bone surface occupied by CTSK+ cells. The total distance between the oral, crest of the bone, and the perpendicular palatine process divided into thirds. The distance occupied by CTSK+ cells was normalized to the length. (B) When measuring the percentage of the total surface occupied by CTSK there was no significant difference (unpaired t test). (C) In controls, the middle and aboral thirds had significantly more distance covered by CTSK than the oral third did. In regions with selective tooth removal, there was a significant decrease in the number of CTSK+ surface in the middle and aboral thirds but no change in the oral third. Two-way analysis of variance and the Benjamini, Krieger, and Yekutieli false discovery rate was used for multiple comparisons. (D) Schematic of experimental design in which BLZ945-soaked beads (20 mg/mL) were placed on one side and DMSO beads were placed on the contralateral side followed by tooth removals at 24 h. (E) Schematic showing the location of beads relative to developing teeth. (F, G) Quantitative reverse transcription polymerase chain reaction result showing significantly less CSF1R and CTSK expression after BLZ945 treatment (unpaired t test). ab, aboral; la, labial; li, lingual; mid, middle; o, oral. Error bars in B, C, F, and G are the ±95% confidence interval.

Local Delivery of BLZ945, a CSF1R Antagonist, Decreases in the Expression of CSF1R and CTSK in Developing Teeth

To test an early signal that may be involved in the induction of odontoclasts, we blocked CSF1, a selective CSF1R antagonist (BLZ945) using bead implants (Fig. 4D; Appendix Fig. 9A–C′). Control beads soaked in a fluorescent dye—CM-DiI—confirmed that dye is released locally in the tissues (Appendix Fig. 9A–C′). Once attached to cell membranes, the dye does not spread unless via cell migration or cell division. BLZ945 exposure for 24 h caused a significant decrease in CSF1R and CTSK expression in developing tooth samples (Fig. 4E–G; Appendix Fig. 9E, F) while other potential target genes were not affected (CSF1, DCSTAMP, ACP5; Appendix Fig. 10A, C). Gene expression returned to control levels by 72 h (Appendix Figs. 9E, F; 10A, B). The decrease in the expression of CSF1R may be due to decreased survival of CSF1R+ cells, as shown in mice treated with other CSF1R antagonists (Rosin et al. 2018).

These results show that CSF signaling is upstream of CTSK and that local release of drugs around developing teeth is sufficient to measure gene expression changes. These methods can be used to build a hierarchy of signaling responses.

Discussion

Here, we characterize the differences between functional and developing teeth using histological and molecular analyses. The novel findings include a full transcriptome of gecko teeth in 2 stages of the life cycle and includes the first sequencing data on teeth undergoing physiological tooth resorption. We defined 3 populations of odontoclasts by their morphology, size, and position. These populations express TRAP and CTSK and may represent 2 stages of odontoclast precursors and the functional cell type. We also tested the relationship between developing and functional teeth and showed that the developing teeth maintain the presence of odontoclast precursors near or attached to functional teeth. We also showed that one of the signals that positively regulates CTSK expression in gecko teeth in vivo is CSF1.

The Developing Teeth and the Relationship to Resorption of Functional Teeth

The mechanisms to shed teeth evolved in conjunction with polyphyodonty. In animals with one set of teeth such as the mouse or rat, odontoclasts are not normally present except in a limited time window (Tsuchiya et al. 2008). In diphyodont mammals, there is a transient appearance of odontoclasts that attach to the roots of primary teeth. The absence of the permanent tooth bud delays the process of tooth shedding (Obersztyn 1963; Lin et al. 2012). In a classic study carried out on dogs, it is clear that the primary tooth roots remain fully intact for 5 mo after the coronal or full follicle is removed (Marks and Cahill 1987). On the other hand, if the coronal follicle is left in place, the primary tooth is never present in the image, implying that the tooth has been resorbed and shed. Similarly, in humans where there is agenesis of premolars resulting in an absent follicle, there is often prolonged retention of the primary molars into adulthood (Bjerklin et al. 2008).

The dental follicle in diphyodont mammals is critical for root resorption. There are data on the expression of genes in mouse dental follicle (Sfrp2, Postn, Aldh1a2, Acta2, Tagln, Gdf10, Cldn2, and Igfbp5) (Krivanek et al. 2020). Thus far, we have identified SFRP2 expression in the mesenchyme surrounding cap-stage teeth and the successional lamina. We removed SFRP2-expressing cells along with the extracted developing teeth, which correlated with a reduction in CTSK+ cells in the middle and aboral thirds. The removal of developing gecko teeth plus attached cells likely delays resorption of functional teeth (Brink et al. 2021).

Commonalities and Differences between Odontoclasts and Osteoclasts

The 2 strengths of the gecko model are that gene expression is conserved with mammals, but unlike mammals such as the mouse and rat, geckos have a steady supply of odontoclasts to study. It makes sense that some features of odontoclasts are the same as osteoclasts based on their likely common origin in fish (Chen et al. 2016). However, in a landmark study that clearly separates observations on odontoclasts from osteoclasts, orthodontic tooth movement induced odontoclasts and root resorption (Ogawa et al. 2019). These authors used a germ-line knockout of Tnfα and treated these animals with orthodontic appliances. They showed that Tnfα is required for odontoclast and osteoclast induction (Ogawa et al. 2019). We extend these findings and show that CSF1 is upstream of CTSK in the odontoclast differentiation pathway similar to osteoclasts (McDonald et al. 2021).

Other studies imply that odontoclasts have separate properties from osteoclasts. Strong evidence from a bone marrow transplant experiment in mice showed that odontoclasts have different origins than osteoclasts. Bone marrow contains osteoclast progenitor cells (McDonald et al. 2021). Mice injected with bone marrow from a donor knockout animal lacking Tnfα showed that odontoclasts were still present in the chimeric animal even though osteoclasts were lacking (Ogawa et al. 2019). These data strongly suggest that the origins of odontoclasts are from other tissues in the host, such as the connective tissue stroma (Ogawa et al. 2019). These data fit our observations that CTSK+ cells seem to be resident apical functional teeth.

Another line of evidence for osteoclasts and odontoclasts not being the same cell type comes from attempts to block root resorption using drugs that are used to treat osteoporosis in animal models. In a model in which root resorption is induced by extracting teeth and replanting them, topical treatment of the roots with bisphosphonates did not prevent resorption (Najeeb et al. 2017). Different results were obtained in a model of orthodontic-induced root resorption in which root resorption was inhibited at the same time as the inhibition of bone remodeling (Liu et al. 2004). Studies using denosumab, a monoclonal antibody that blocks RANK, have reported that there is increased risk of root resorption in patients. A retrospective study consisting of patients taking denosumab without taking bisphosphonates (Alyahya and Myers 2021) showed an association with increased root resorption; however, the authors admitted that the exact timing of drug therapy was unknown, so they could not rule out a rebound effect (Deeb et al. 2019). In a case report, a patient actively receiving denosumab therapy that successfully treated the osteoporosis developed rampant, external cervical resorption with no other risk factors (Mikuskova et al. 2022). Therefore, there is a need to test denosumab further in a model such as the gecko to determine whether there is a separate, RANK-independent pathway that regulates odontoclast function. Other proteolytic enzymes identified in this study should also be examined to determine whether they are expressed in odontoclasts and whether they mediate odontoclast root resorption.

Correlation with Human Genes Associated with Overretention of Deciduous Teeth

We found clinical correlates of the genes differentially expressed in functional teeth with genes that cause osteopetrosis, a bone disease characterized by increased bone density due to impaired osteoclast differentiation or function. We were intrigued to find that RNA-seq profiling of functional teeth included 6 genes out of the 9 that cause autosomal recessive osteopetrosis (OSTM1, CA2, TC1RG1, TNFSF11, CLCN7, TNFRSF11A; Table 1). We hypothesized that if the genes were expressed in odontoclasts, then patients with osteopetrosis would have overretained deciduous teeth. Indeed, osteopetrosis type 2 (OMIM 259710) caused by variants of TNFSF11 have overretained primary teeth (Kahler et al. 1984). All forms of osteopetrosis have dental eruption problems, but not all were diagnosed as a specific failure to resorb the primary teeth. More careful dental phenotyping in children is needed. We have used CTSK+ as a marker of odontoclasts in this study. Human variants in CTSK cause pycnodysostosis (OMIM #265800), which often includes a double row of teeth due to persistence of primary teeth (Edelson et al. 1992). Thus, we have 2 promising approaches for future work: targeting the initial induction of odontoclasts and inhibiting the function of the differentiated odontoclasts. Taken together, the human correlates and conserved gene expression profiles support the continued use of the gecko for preclinical odontoclast research.

Here, we identified a panel of proteolysis enzymes that correlate with higher numbers of odontoclasts in functional teeth. Most of these enzymes are either not found in osteoclasts or are expressed at low levels, some of which may be more adapted to degrading dentin matrix. Dentin matrix has proportionately higher levels of dentin matrix protein (DMP1) and dentin sialoprotein (DSP) compared with bone (Butler et al. 2003). Osteopontin is another noncollagenous glycoprotein in dentin that is required for odontoclast but not osteoclast formation in response to orthodontic tooth movement (Chung et al. 2008).

Author Contributions

J.I. Henriquez, J.M. Richman, contributed to conception, design, data acquisition, analysis, and interpretation, drafted and critically revised the manuscript; S. Flibotte, contributed to data analysis, critically revised the manuscript; K. Fu, contributed to data acquisition and analysis, critically revised the manuscript. All authors gave final approval and agree to be accountable for all aspects of the work.

Supplemental Material

sj-pdf-1-jdr-10.1177_00220345241304756 – Supplemental material for Molecular Profiling of Odontoclasts during Physiological Tooth Replacement

Supplemental material, sj-pdf-1-jdr-10.1177_00220345241304756 for Molecular Profiling of Odontoclasts during Physiological Tooth Replacement by J.I. Henriquez, S. Flibotte, K. Fu and J.M. Richman in Journal of Dental Research

Research Data 1

sj-xlsx-2-jdr-10.1177_00220345241304756 – Supplemental material for Molecular Profiling of Odontoclasts during Physiological Tooth Replacement

sj-xlsx-2-jdr-10.1177_00220345241304756 for Molecular Profiling of Odontoclasts during Physiological Tooth Replacement by J.I. Henriquez, S. Flibotte, K. Fu and J.M. Richman in Journal of Dental Research

This article is distributed under the terms of the Creative Commons Attribution 4.0 License (http://www.creativecommons.org/licenses/by/4.0/) which permits any use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage).

Research Data 2

sj-xlsx-3-jdr-10.1177_00220345241304756 – Supplemental material for Molecular Profiling of Odontoclasts during Physiological Tooth Replacement

sj-xlsx-3-jdr-10.1177_00220345241304756 for Molecular Profiling of Odontoclasts during Physiological Tooth Replacement by J.I. Henriquez, S. Flibotte, K. Fu and J.M. Richman in Journal of Dental Research

Footnotes

Acknowledgements

We thank Dr. P. Panwar for technical advice on TRAP staining and quantification of odontoclasts.

Correction (February 2025):

Article updated to correct textual errors and replace Figure 3 with a revised version. See for more details.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Natural Sciences and Engineering Research Council of Canada grants RGPIN-2016-05477 and RGPIN-2023-05145 and National Institutes of Health grant 5R21DE026839-02 to JMR.

ORCID iD

J.M. Richman

A supplemental appendix to this article is available online.

References

Alyahya

Myers

. 2021. Denosumab use as a predictor variable for external cervical resorption: a case-control study. J Endod. 47(3):366–373.

Bertin

TJC

Thivichon-Prince

LeBlanc

ARH

Caldwell

Viriot

. 2018. Current perspectives on tooth implantation, attachment, and replacement in amniota. Front Physiol. 9:1630.

Bjerklin

Al-Najjar

Karestedt

Andren

2008. Agenesis of mandibular second premolars with retained primary molars: a longitudinal radiographic study of 99 subjects from 12 years of age to adulthood. Eur J Orthod. 30(3):254–261.

Brink

Henriquez

Grieco

Martin Del Campo

Richman

. 2021. Tooth removal in the leopard gecko and the de novo formation of replacement teeth. Front Physiol. 12:576816.

Brink

Chuong

Richman

. 2020. The effects of premature tooth extraction and damage on replacement timing in the green iguana. Integr Comp Biol. 60(3):581–593.

Butler

Brunn

Qin

2003. Dentin extracellular matrix (ECM) proteins: comparison to bone ECM and contribution to dynamics of dentinogenesis. Connect Tissue Res. 44(suppl 1):171–178.

Chen

Blom

Sanchez

Tafforeau

Ahlberg

. 2016. The stem osteichthyan andreolepis and the origin of tooth replacement. Nature. 539(7628):237–241.

Chung

Soma

Rittling

Denhardt

Hayata

Nakashima

Ezura

Noda

2008. OPN deficiency suppresses appearance of odontoclastic cells and resorption of the tooth root induced by experimental force application. J Cell Physiol. 214(3):614–620.

Deeb

Azarnoush

Laskin

Deeb

. 2019. Discontinuation of denosumab as a potential cause of generalized external cervical root resorption: a case report. J Endod. 45(5):640–644.

10.

Domon

Sugaya

Yawaka

Osanai

Hanaizumi

Takahashi

Wakita

1994. Electron microscopic and histochemical studies of the mononuclear odontoclast of the human. Anat Rec. 240(1):42–51.

11.

Drake

Dodds

James

Connor

Debouck

Richardson

Lee-Rykaczewski

Coleman

Rieman

Barthlow

, et al. 1996. Cathepsin K, but not cathepsins B, L, or S, is abundantly expressed in human osteoclasts. J Biol Chem. 271(21):12511–12516.

12.

Edelson

Obad

Geiger

Artul

HJ.

1992. Pycnodysostosis. Orthopedic aspects with a description of 14 new cases. Clin Orthop Relat Res. (280):263–276.

13.

Fuenzalida

Illanes

Lemus

Guerrero

Oyarzun

Acuna

Lemus

1999. Microscopic and histochemical study of odontoclasts in physiologic resorption of teeth of the polyphyodont lizard, Liolaemus gravenhorsti. J Morphol. 242(3):295–309.

14.

Handrigan

Leung

Richman

. 2010. Identification of putative dental epithelial stem cells in a lizard with life-long tooth replacement. Development. 137(21):3545–3549.

15.

Handrigan

Richman

. 2010. A network of Wnt, hedgehog and BMP signaling pathways regulates tooth replacement in snakes. Dev Biol. 348:130–141.

16.

Henriquez

Richman

. 2024. Resilience of the replacing dentition in adult reptiles. Dev Biol. 516:71–81.

17.

Iglesias-Linares

Hartsfield

Jr.

2017. Cellular and molecular pathways leading to external root resorption. J Dent Res. 96(2):145–152.

18.

Kahler

Burns

Aylsworth

. 1984. A mild autosomal recessive form of osteopetrosis. Am J Med Genet. 17(2):451–464.

19.

Krivanek

Soldatov

Kastriti

Chontorotzea

Herdina

Petersen

Szarowska

Landova

Matejova

Holla

, et al. 2020. Dental cell type atlas reveals stem and differentiated cell types in mouse and human teeth. Nat Commun. 11(1):4816.

20.

LeBlanc

ARH

Palci

Anthwal

Tucker

Araujo

Pereira

MFC

Caldwell

. 2023. A conserved tooth resorption mechanism in modern and fossil snakes. Nat Commun. 14(1):742.

21.

Lin

Zhao

Yang

. 2012. Root resorption of primary molars without successor teeth. An experimental study in the beagle dog. Eur J Oral Sci. 120(2):147–152.

22.

Liu

Igarashi

Haruyama

Saeki

Shinoda

Mitani

2004. Effects of local administration of clodronate on orthodontic tooth movement and root resorption in rats. Eur J Orthod. 26(5):469–473.

23.

Marks

Jr Cahill

. 1987. Regional control by the dental follicle of alterations in alveolar bone metabolism during tooth eruption. J Oral Pathol. 16(4):164–169.

24.

McDonald

Kim

Mulholland

Rauner

2021. New insights into osteoclast biology. JBMR Plus. 5(9):e10539.

25.

Mikuskova

Vanuga

Adamicova

Statelova

Janickova

Malachovsky

Siebert

2022. Multiple idiopathic external cervical root resorption in patient treated continuously with denosumab: a case report. BMC Oral Health. 22(1):129.

26.

Najeeb

Siddiqui

Khurshid

Zohaib

Zafar

Ansari

. 2017. Effect of bisphosphonates on root resorption after tooth replantation—a systematic review. Dent Traumatol. 33(2):77–83.

27.

Obersztyn

1963. Experimental investigation of factors causing resorption of deciduous teeth. J Dent Res. 42:660–674.

28.

Ogawa

Kitaura

Kishikawa

Shen

Ohori

Noguchi

Marahleh

Nara

Ochi

, et al. 2019. Tnf-alpha is responsible for the contribution of stromal cells to osteoclast and odontoclast formation during orthodontic tooth movement. PLoS One. 14(10):e0223989.

29.

Philbrick

Dreyer

Nakchbandi

Karaplis

. 1998. Parathyroid hormone-related protein is required for tooth eruption. Proc Natl Acad Sci U S A. 95(20):11846–11851.

30.

Rosin

Vora

Kurrasch

. 2018. Depletion of embryonic microglia using the csf1r inhibitor plx5622 has adverse sex-specific effects on mice, including accelerated weight gain, hyperactivity and anxiolytic-like behaviour. Brain Behav Immun. 73:682–697.

31.

Sahara

Toyoki

Ashizawa

Deguchi

Suzuki

1996. Cytodifferentiation of the odontoclast prior to the shedding of human deciduous teeth: An ultrastructural and cytochemical study. Anat Rec. 244(1):33–49.

32.

Salomies

Eymann

Khan

Di-Poi

2019. The alternative regenerative strategy of bearded dragon unveils the key processes underlying vertebrate tooth renewal. Elife. 8:e47702.

33.

Smyth

Sasiwachirangkul

Williams

Scott

. 2022. Cathepsin s (CTSS) activity in health and disease—a treasure trove of untapped clinical potential. Mol Aspects Med. 88:101106.

34.

Takada

Kajiya

Fukushima

Okamoto

Motokawa

Okabe

2004. Calcitonin in human odontoclasts regulates root resorption activity via protein kinase a. J Bone Miner Metab. 22(1):12–18.

35.

Tsai

Abdelhamid

Khan

Elkarargy

Widelitz

Chuong

2016. The molecular circuit regulating tooth development in crocodilians. J Dent Res. 95(13):1501–1510.

36.

Tsuchiya

Akiba

Takahashi

Sasano

Kashiwazaki

Tsuchiya

Watanabe

2008. Comparison of expression patterns of cathepsin K and MMP-9 in odontoclasts and osteoclasts in physiological root resorption in the rat molar. Arch Histol Cytol. 71(2):89–100.

37.

Tsukasaki

Huynh

Okamoto

Muro

Terashima

Kurikawa

Komatsu

Pluemsakunthai

Nitta

Abe

, et al. 2020. Stepwise cell fate decision pathways during osteoclastogenesis at single-cell resolution. Nat Metab. 2(12):1382–1390.

38.

Vonk

Zhao

Pan

Hudnall

Oakes

Lopez

Hasel-Kolossa

Kuncz

AWC

Sengelmann

Gamble

, et al. 2023. Single-cell analysis of lizard blastema fibroblasts reveals phagocyte-dependent activation of hedgehog-responsive chondrogenesis. Nat Commun. 14(1):4489.

39.

Wang

Zhao

Wang

Jiang

Zhao

Qing

2023. Cathepsin H: Molecular characteristics and clues to function and mechanism. Biochem Pharmacol. 212:115585.

40.

Woerdeman

1921. Beitrage zur entwicklungsgeschichte von zahnen und gebisse der reptilien. Beitrag iv: Uber die anlage und entwicklung der zahne. Archiv fur mikroskopische Anatomie. 95:265–395.

41.

Jiang

T-X

Elsey

Temple

Divers

Glenn

Yuan

Cheng

M-H

Widelitz

, et al. 2013. Specialized stem cell niche enables repetitive renewal of alligator teeth. Proc Natl Acad Sci U S A. 110(22):E2009–E2018.

42.

Xiao

Qian

Wang

2022. Advances in the study of the mechanisms of physiological root resorption in deciduous teeth. Front Pediatr. 10:850826.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

3.30 MB

0.44 MB

0.03 MB