The Effects of Premature Tooth Extraction and Damage on Replacement Timing in the Green Iguana

Synopsis

Reptiles with continuous tooth replacement, or polyphyodonty, replace their teeth in predictable, well-timed waves in alternating tooth positions around the mouth. This process is thought to occur irrespective of tooth wear or breakage. In this study, we aimed to determine if damage to teeth and premature tooth extraction affects tooth replacement timing long-term in juvenile green iguanas (Iguana iguana). First, we examined normal tooth development histologically using a BrdU pulse-chase analysis to detect label-retaining cells in replacement teeth and dental tissues. Next, we performed tooth extraction experiments for characterization of dental tissues after functional tooth (FT) extraction, including proliferation and β-Catenin expression, for up to 12 weeks. We then compared these results to a newly analyzed historical dataset of X-rays collected up to 7 months after FT damage and extraction in the green iguana. Results show that proliferation in the dental and successional lamina (SL) does not change after extraction of the FT, and proliferation occurs in the SL only when a tooth differentiates. Damage to an FT crown does not affect the timing of the tooth replacement cycle, however, complete extraction shifts the replacement cycle ahead by 4 weeks by removing the need for resorption of the FT. These results suggest that traumatic FT loss affects the timing of the replacement cycle at that one position, which may have implications for tooth replacement patterning around the entire mouth.

Introduction

The ability for animals to continuously replace their teeth, or polyphyodonty, is deeply nested within Vertebrata (Edmund 1960). Detailed surveys of tooth replacement in nonmammalian vertebrates, including teleost fish (Berkovitz and Moore 1974, 1975), crocodilians (Edmund 1962; Kieser et al. 1993), dinosaurs (Schwarz et al. 2015; Hanai and Tsuihiji 2019; He et al. 2018), pterosaurs (Fastnacht 2008), and lizards (Edmund 1960, 1969; Cooper 1966; Kline and Cullum 1984; Kline and Cullum 1985; Grieco and Richman 2018) revealed that teeth are replaced in alternating tooth positions in apparent waves around the mouth. Some potential mechanisms for the initiation of this pattern have been proposed relating to signals and tissue growth in the dental lamina (DL), the tissue responsible for forming teeth (Edmund 1969; Osborn 1970, 1971; Westergaard and Ferguson 1987; Murray and Kulesa 1996; Fraser et al. 2006a; Huysseune and Witten 2006; Lamoureux et al. 2018; Gibert et al. 2019; Sadier et al. 2020). However, less is known about how cyclical patterns of tooth replacement are maintained later in life as juveniles and adults, and if traumatic tooth loss or damage would disrupt the natural timing of tooth replacement.

Previous investigations into the mechanisms of tooth renewal have shown that epithelial tissue with odontogenic potential is retained throughout life, either a DL with successional lamina (SL) in squamates (lizards, snakes, and amphisbaenids), sharks, and crocodilians (Handrigan and Richman 2010a; Jernvall and Thesleff 2012; Fraser et al. 2013; Wu et al. 2013; Tucker and Fraser 2014; Tsai et al. 2016; Salomies et al. 2019), or as a middle dental epithelium without SL in salmon and bichir (Vandenplas et al. 2016). The DL is thought to be the source of putative stem cells required for tooth renewal based on the position of slow-cycling label retaining-cells (Huysseune and Thesleff 2004; Smith et al. 2009; Handrigan et al. 2010).

Tooth renewal in reptiles happens in a highly patterned, well-timed manner (Edmund 1962, 1969; Cooper 1966; Kline and Cullum 1984; Kline and Cullum 1985; Grieco and Richman 2018). In Edmund’s (1960) survey of nonmammalian vertebrates, he stated that tooth replacement is not affected by damage to a tooth, and that replacement is controlled by unknown intrinsic factors. Tooth damage and wear are common in larger reptiles, such as dinosaurs (Farlow and Brinkman 1994; Schubert and Ungar 2005) and alligators, where injuries occur during feeding and gaps along the toothrow and retention of highly worn teeth in wild and captive animals are common (Hall 1985; Erickson 1996). A study of tooth replacement in farmed salmon has shown that damage to a functional tooth (FT) does not result in faster resorption or replacement (Huysseune et al. 2012). In contrast, Wu et al. (2013) found that premature removal of teeth in the alligator led to an increase in stem cell proliferation. These early changes in cellular dynamics may lead to more rapid tooth initiation; however, longer follow-up is needed to examine how this affects the timing of the tooth replacement cycle. Therefore, it is not known if damage or extraction of a tooth can stimulate stem cells to renew a tooth more quickly and increase the rate of tooth replacement to be faster than would be expected naturally.

Crocodilians have a different arrangement of the FT in relation to the DL than squamates. In adult crocodilians, the teeth are in individual sockets and the DL is a small remnant of epithelium, separate from the FT (Wu et al. 2013). In adult squamates, the DL retains a connection with the oral epithelium, and teeth remain connected to the DL throughout development (Richman and Handrigan 2011). Thus, the response to tooth removal in crocodilians may be different in reptiles with a continuous connection between functional and replacement teeth. Inductive signals could be transmitted along the DL, perhaps initiating a new tooth bud earlier than normal. The reverse could also be true, where removing an FT severs the DL which delays successional tooth formation.

During the course of Edmund’s (1960, 1962, 1969) research at the Royal Ontario Museum (ROM), he collected longitudinal data from >30 species of reptiles and amphibians for up to 2 years using X-rays. Some of this dataset was published for alligator (Edmund 1962) and the green iguana (Edmund 1969), where he showed that replacement occurs in waves passing from the back to the front of the mouth. However, he also carried out tooth removal experiments in young green iguanas. Functional teeth were removed, or the crowns were broken, and regeneration and patterning were followed for up to 7 months post-extraction. Although the original intent behind these experiments is unknown, the data can be examined now in light of a modern understanding of tooth renewal in reptiles, as this dataset offers a rare opportunity to examine tooth replacement rates longitudinally following disruption. Patterning and replacement time in nontreated green iguanas are relatively well-known, as they have been the subject of other studies on longitudinal tooth replacement patterns (Kline 1983; Kline and Cullum 1984, 1985) and tooth development (Howes 1979; Leche 1893). The results of these studies determined that tooth shedding occurs 5 to 6 times/year, with a shed-to-shed tooth cycle of ∼10 weeks. A tooth position is empty for no longer than 2 weeks, and teeth are added at the back of the tooth row as the animal grows (Howes 1979; Kline and Cullum 1984; Kline and Cullum 1985). Additionally, there is no significant difference in tooth size between the mandibular and maxillary teeth (Kline 1983).

In order to supplement our analysis of the historical longitudinal data, a series of tooth extraction surgeries were recently performed in green iguanas in order to examine the histology and development of teeth following disruption. The combination of Edmund’s longitudinal data with new data on tooth extraction in green iguanas will test whether or not damage to the teeth or premature tooth extraction results in faster tooth replacement in a representative squamate.

Methods

Animal husbandry for tooth extraction experiments

Twelve juvenile (∼1 year old, snout–vent lengths ranging from 17 to 22 cm, 380–500 g) green iguanas (Iguana iguana) were obtained from local vendors in California, USA. Animals were housed in the University of Southern California (USC) animal facility. Animals were kept on a cycle of 10 h light at 28°C and 14 h dark at 23°C. Iguanas were fed a diet of green vegetables and Natural Adult Iguana Food Pellets (Zoo Med, Costa Mesa, CA, USA). All procedures were approved by the USC Institutional Animal Care and Use Committee.

Histological examination of a normal tooth cycle

Two iguanas were used for histological examination of normal tooth development. After euthanization, frontal sections through the maxilla and mandible were prepared.

Examination of tooth regeneration

Five iguanas each had three nonadjacent functional teeth extracted from the left maxilla or mandible, and the right sides were used as controls (n = 15 extracted teeth). Prior to extraction, animals were anesthetized with Ketamine (20 mg/kg) and Xylazine (2 mg/kg). Forceps were used to probe the functional teeth. If an FT was loose, it was not plucked since the base had already been fully resorbed. Only firmly attached FT were removed by pushing them toward the lingual side and breaking the connection of the tooth with the bone. Pain was relieved by administering ketoprofen (2 mg/kg). Iguanas were euthanized after 1 week (n = 1), 3 weeks (n = 2), 6 weeks (n = 1), and 12 weeks (n = 1). After euthanasia, we confirmed that the SL was undifferentiated or had recently differentiated to a cap stage tooth with histology before using the tissue for further analysis.

BrdU pulse study

To detect transit amplifying (TA; rapidly proliferating) cells in the iguana DL and SL, we used a short-term BrdU pulse labeling method. All five iguanas with extracted teeth were injected with BrdU (Sigma) intraperitoneally (50 mg/kg) 3 h prior to euthanasia.

BrdU pulse-chase label retention study

To detect slow cycling, label-retaining cells (LRCs) in the normal iguana dentition, animals were injected with BrdU (50 mg/kg) twice a day for 1 week and samples were collected at 0 (n = 1), 2 (n = 2), and 8 weeks (n = 2). This method has been widely used to detect putative stem cells in hair follicles (Cotsarelis et al. 1990), feather follicles (Yue et al. 2005), alligator scales (Wu et al. 2018), and reptile dental tissues (Handrigan et al. 2010; Wu et al. 2013; Salomies et al. 2019). The principle is to continuously label all fast cycling (TA) and slow cycling (label retaining) cells, then wait for a certain period of time (the chase). After the chase, fast cycling cells lose their BrdU label, whereas slow-cycling cells (the putative stem cells) retain the label.

Paraffin sections

The jaws of euthanized iguanas were decalcified with 0.5 M  Ethylenediaminetetraacetic acid (EDTA) for 3 days to 2 weeks after fixation in paraformaldehyde. For each sample, continuous paraffin sections (10  μm frontal sections) covering at least 10 tooth families were prepared. The sections covered the sites where teeth were extracted and control areas. Hematoxylin and eosin (H&E) staining was performed for every 18 sections. The stage of development of each tooth family unit was ascertained by the structure of DL and SL. Sections from untreated tooth families were used as control tissue.

Immunohistochemistry

We performed immunostaining to detect the distribution of β-Catenin (an indication of Wnt signaling) and Tenascin-C (a neural cell adhesion molecule) in the iguana tooth family unit. In the alligator DL, nuclear β-Catenin was co-localized with LRCs in the DL “bulge,” and Tenascin-C was used to detect the stroma surrounding the dental tissues (Wu et al. 2013). Immunostaining was performed as described in Wu et al. (2013). The monoclonal β-catenin antibody is from Sigma-Aldrich (C7207). The BrdU antibody is from BD Biosciences (347580). The Antibody to tenascin-C is the same as previously described in (Chuong and Chen 1991).

X-ray analysis

An historical dataset collected in the 1960s by Gordon Edmund of the ROM was re-analyzed to observe changes in tooth replacement patterning and timing after tooth removal and damage. Details on how the radiographs were taken were described by Edmund (1960): “With modern fine-grain industrial films (Kodak type AA or M), a small focal spot, and a long tube-to-film distance, excellent detail can be obtained, permitting much enlargement, at least up to thirty diameters. Radiography was also used for periodic examination of live animals.” Details of the surgeries were taken from Edmund’s lab notes, housed at the ROM. Three animals were sedated with IP injections of Nembutal before surgery and before every X-ray image was taken. Four functional adjacent teeth, regardless of the state of development of each functional or replacement tooth, were extracted from the middle of the left maxilla in animal R309, 4 from the middle of the left mandible of animal R305, and 6 from the middle of the right maxilla of animal R349 (Table 1). Replacement teeth and soft tissues were not touched, so tooth bases were left behind when the crowns were snapped off. The iguanas were X-rayed monthly (R309 and R305) and bi-monthly (R349) over a period of up to seven months before and after the surgery. The ages of the animals were not known, however had snout-vent lengths of 15 cm (R305 and R309) and 19.6 cm (R349) and weighed 60–90g (R305 and R309) and 180–220g (R349) throughout the course of the experiment, which is consistent with previous reports of weights in 0–3 year old iguanas (Kline and Cullum 1984).

Table 1.

Time (months) between shedding events at each tooth position in the green iguana

	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19	20	21	Avg
R305 Lmd	5.0	5.5	5.5	6.0	4.0	3.7	5.5	5.5	5.0	6.0	5.0	6.0	6.0	5.0	5.5	5.0	5.5	3.3	5.5			5.2
R309 Lmx	7.5	3.5	4.0	4.3	4.5	4.7	4.3	3.0	4.3	4.0	7.0	4.7	4.0	4.0	4.5	4.3						4.5
R349 Rmx	4.5	5.0	5.0	5.0	4.5	5.0	5.0	5.0	5.0	5.0	5.0	5.0	4.5	4.5	5.0	5.0	4.5	4.5	4.5	4.5	6.0	4.9

Gray boxes indicate surgery areas.

X-rays were digitized using a light table and digital camera and scored manually using a grid, following Grieco and Richman (2018). Images of the X-rays from the 3 surgery animals are available online through Dataverse (Brink 2020). Time between shedding events (tooth loss) was counted along the jaw to see if there were any differences between tooth replacement timing before the surgery and after the surgery, and in control areas of the mouth and the surgical areas of the mouth. There was some inconsistency in the timing of X-ray collection (different number of days between each time an X-ray was taken), and so not every shedding event was captured on the date the X-ray was taken. Therefore, teeth were scored as shed on the date the X-ray was taken, even if the tooth had been shed a week or two prior and a new tooth had already erupted, based on a series of factors. Newly erupted teeth have thin dentine walls in the tooth base, no visible replacement tooth, less opaque dentine than neighboring teeth, and crowns that appear to be ‘floating’ in the socket (no basal development). In ROM R309, the last three tooth positions (17–19) were not counted due to too much missing data and the addition of teeth to the toothrow. Shedding events in treated and non-treated areas were compared using a student’s T-test.

Results

Dental morphology and normal histology

A small amount of skeletal material from Edmund’s experiments is housed at the ROM (Fig. 1A and B) for macroscopic analysis of tooth morphology and attachment. The functional teeth have a pleurodont attachment to the jaw and the lingual surfaces of the teeth are visible inside the mouth (Fig. 1B), as is typical for reptiles (Bertin et al. 2018). H&E staining of paraffin sections from new animals confirms that the iguana tooth family includes an FT, an RT, DL, and SL (Fig. 1C, C′, D, and D′). The DL extends from the oral epithelium down below the lingual side of the FT (Fig. 1C and D). The SL is the free end of the DL, and is present lingual to the developing RT (Fig. 1D′). As the RT develops from the differentiation of the SL, resorption of the base of the FT begins (Fig. 1C′) creating large resorption pits in the FT (Fig. 1B and D). The replacement teeth fell out of Edmund’s iguana skull during the skeletonization process (the removal of soft tissue by insect activity), as they are held in place with soft tissue only during life (Fig. 1B). Only two teeth are ever present at one time at each tooth position. Sections from a nontreated animal through three adjacent teeth show the morphology of the DL in interdental and dental positions (Supplementary Fig. S1). The DL is thicker where it remains attached to developing and functional teeth and is thinner farther away from the neighboring teeth, depending on the angle of section (Supplementary Fig. S1).

Fig. 1.

Tooth anatomy and histology of I. iguana (color in supplementary files, B&W in print). (A) I. iguana skull in right lateral view, from Edmund’s original experiments, ROM R337. (B) Teeth of I. iguana, with resorption pits in functional teeth for replacement teeth (not preserved when soft tissue removed from skull, arrows). (C–C′) Histology of a tooth family with RT in late cap stage, resorbing FT. (D–D′) Histology of an FT with DL and RT with undifferentiated SL. (E–E′) β-Catenin staining. (F–F′) Tn-C staining. (G–G′) Staining of a 3-h BrdU pulse. (H–H′) Staining of a 1-week BrdU pulse labeling (Time 0, no chase) with positive cells (arrows) in the dl (dotted outline). (I–I′) Staining for BrdU after a 2-week chase. (J–J′) Staining for BrdU after an 8-week chase. (K) Diagram illustrates the distribution of TA cells (black dots) and LRCs (white dots) in the iguana DL. (L) Counts of BrdU-positive cells in the DL and SL at 1 week (no chase), 2 weeks chase, and 8 weeks chase. N = 3 teeth/timepoint. Scale bar: (A) 1 cm, (B) 1 mm, (C–J) 100 µM.

The immunohistochemical experiments detected β-Catenin expression in the enamel organ surrounding the RT and the DL, however, only diffuse expression was detected in the SL (Fig. 1E and E′). Tenascin-C (Tn) was detected in the stroma surrounding the SL (Fig. 1F and F′).

A pulse of BrdU administered 3 h before euthanasia detected rapidly proliferating TA cells in the cervical loops of the RT, and very few BrdU positive cells were detected in the SL and DL (Fig. 1G and G′ arrows).

The BrdU pulse-chase experiment detected LRCs or putative stem cells in normally cycling iguana dental tissues. Iguanas were labeled for 1 week and chased for 0 , 2 , and 8 weeks (Fig. 1H–J′). At Time 0 (no chase), BrdU-positive cells were detected in the labial side of the DL (Fig. 1H′, arrows) and relatively few cells are labeled in the SL (Fig. 1H′). After 2 weeks chase, the LRCs are scarcely detected and mainly distributed along the lingual margin of the DL (Fig. 1I and I′, arrows), and relatively few labeled cells are detected in the SL. After 8 weeks chase, the LRCs are more dilute but still detectable and they are randomly distributed in the DL (Fig. 1J), and relatively few labeled cells are detected in the SL (Fig. 1J′). We show three sections comprising 300 µm of tissue through one tooth family after 1 week (no chase) and 8 weeks chase (Supplementary Fig. S2). These examples demonstrate the consistent lack of BrdU positive cells in the SL. We summarized the distribution of TA cells (black dots) and LRCs (white dots) in the iguana DL in Fig. 1K. Counts of BrdU positive cells in the DL and SL are compared at 1 week (Time 0), 2 weeks, and 8 weeks in Fig. 1L.

Analysis of tooth regeneration following extraction

One week following FT extraction, the oral epithelium is dysmorphic (Fig. 2A), but the RT does not appear to be affected and is developing normally (Fig. 2A′). Proliferative cells are present in the cervical loops after a 3-h pulse of BrdU (Fig. 2B). There are no BrdU-postitive cells in the SL (Fig. 2B). However, there is nuclear β-catenin expressed in the SL and cervical loops of the developing tooth, suggesting activation of canonical Wnt signaling (Fig. 2C and C′).

Fig. 2.

Differentiation of DL after FT extraction in I. iguana (Color in supplementary files, B&W in print). (A–L) Tooth regeneration 1 (A–C′), 3 (D–F′), 6 (G–I′), and 12 weeks (J–L) after FT extraction. (A–A′, D–D′, G–G′, and J–J′) H&E staining. There is no change in morphology in the SL until 12 weeks when it differentiates to a tooth bud. (B, E, H, and K) Staining of a 3-h BrdU pulse labeling. The SL does not have BrdU-positive cells after regenerating for 1(B), 3 (E), and 6 weeks, (H) however, cell proliferation is detected in the cervical loops of the developing RT. Cell proliferation is detected in the SL after 12 weeks of regeneration after differentiation of a new tooth (K, arrows). (C–C′ F–F′, I–I′ and L–L′) β-catenin staining. Nuclear β-catenin staining can be detected in the SL 1, 3, and 6 weeks after FT (arrows). β-catenin is diffusely present at 6 weeks. Nuclear β-catenin staining can be detected in the developing tooth bud at 12 weeks (arrows). Scale bar: 100 µM.

Three and 6 weeks after extraction of the FT, the RT is continuing to develop normally (Fig. 2D, D′, G, and G′). The SL, after a 3-h pulse of BrdU, is non-proliferative at both timepoints (Fig. 2E and H). Nuclear β-Catenin is detected at the 3-week timepoint in the SL and odontoblasts of the developing tooth (Fig. 2F and F′). At 6 weeks, β-Catenin is strongly expressed in the SL but is not nuclear (Fig. 2I and I′).

At 12 weeks post-extraction, a cap-stage tooth is present at the tip of the SL (Fig. 2J and J′), and BrdU-positive cells are present in the epithelium of the newly differentiated tooth (Fig. 2K). Nuclear β-Catenin is present at the developing crown of the new tooth (Fig. 2L and L′). Comparison of treated teeth to a cap-stage tooth from the nontreated (control) tissue shows no differences in the distribution of BrdU positive cells (Supplementary Fig. S3).

X-ray data

We assessed normal tooth replacement rates in the Edmund dataset by examining teeth outside of the surgical treatment site. The tooth replacement cycle varies between animals and between teeth positions, ranging from 2 to 6 months between sheds (Table 1). The average amount of time that passed between tooth sheds at one tooth position in the three individuals is 4.9 and 4.5 months in the maxilla, and 5.2 months in the mandible (Table 1). Therefore, a shed-to-shed cycle is ∼20 weeks in these three animals. No significant difference in timing was noted between treated and nontreated areas of the mouth (Supplementary Figs. S4–S6 and Table 1). All teeth had gone through one or two replacement cycles by the end of the experiment (except for one tooth position in one animal (R349, Fig. 3A′′′; Supplementary Fig. S4) which had been too damaged), meaning that enough time had passed for new teeth to initiate and shed after the extraction took place.

Fig. 3.

X-rays of tooth extraction and recovery in medial view in I. iguana. (A–A′′′) ROM R349, right maxilla. (A) before tooth extraction. Teeth 12–17 were removed during surgery (arrows). (A′) immediately after extraction. Teeth 12, 14, 15, and 17 were incompletely removed (asterisks). (A′′) 1-month post-extraction. Tooth 13 and tooth 16 erupted (arrows). (A′′′) 6 months post-extraction. Tooth position 15 suffered damage and did not return by the end of the experiment (arrow). (B–B′′′) ROM R309, left maxilla. (B) before tooth extraction. Teeth 12–15 were removed during surgery (arrows). (B′) 1-week post-extraction. Tooth 12 was incompletely removed (asterisk). (B′′) 2 months post-extraction. Tooth 13 erupted shorter than teeth in the rest of the toothrow (arrow). (B′′′) 7 months post-extraction. Tooth 13 has been shed and now has a normal phenotype. (C–C′′′) ROM R305, left mandible. (C) before tooth extraction. Teeth 10–13 were removed during surgery (arrows). (C′) 1-week post-extraction. Tooth 10 and 12 were incompletely removed (asterisks). (C′′) 3 months post-extraction. Tooth 11 erupted and ankylosed to the jaw at an angle (arrow). (C′′′) 6 months post-extraction. Teeth have normal phenotypes. Black dots are marks made by A.G. Edmund during preliminary analyses in the 1960s.

Analysis of the X-rays immediately following extraction (Fig. 3A′) and 1-week post-extraction (Fig. 3B′ and C′) shows differences in the extent of tooth removal. Several teeth were broken at the crown, leaving the apical portion of the tooth connected to the jaw in an effort to not disturb the DL, as described in Edmund’s notes (Fig. 3, asterisks). Functional teeth that were fully removed had likely already undergone some resorption, and so complete extraction was easier.

In R349, the right maxilla, six teeth were removed in total, four of which were incompletely removed (Fig. 3A′, asterisks at teeth 12, 14, 15, and 17; Supplementary Fig. S4). One-month post-extraction, teeth 13 and 16 erupted normally (Fig. 3A′′). At 6 months post extraction, all teeth had returned except for position 15 (Fig. 3A′′′). A small, abnormal tooth appears to be lodged sideways at this position, and it is not known what caused it or if it would have shed if the experiment carried on for several more months.

In R309, the left maxilla, four teeth were removed in total (Fig. 3B), one of which was incompletely removed (tooth 12, asterisk in Fig. 3B′). One week after surgery, an RT had already erupted (tooth 14, Fig. 3B ′; Supplementary Fig. S5), suggesting that tooth was nearly mature at the time of FT extraction. At 2 months-post extraction, almost all teeth in the extraction area had erupted normally, except for one tooth that erupted and attached to the jawbone but is shorter than its neighboring teeth (tooth 13, Fig. 3B′′). By 7 months post-extraction, the tooth row had filled in and has a normal phenotype.

In R305, the left mandible, four teeth were removed in total, two of which were incompletely removed (teeth 10 and 12, asterisk in Fig. 3C′; Supplementary Fig. S6). Three months post extraction, tooth 11 erupted and fused to the jaw at an angle, likely because there were no teeth next to it to support it and guide it through the eruption pathway (Fig. 3C′′). After the crooked tooth was shed, the RT at that position and neighboring positions erupted normally at 6 months post-extraction (Fig. 3C′′′).

Discussion

The timing between sheds in the iguanas examined in this study is ∼20 weeks. Given this 20-week shed-to-shed cycle, and using information from histology, a timeline of tooth replacement was deduced (Fig. 4A). The first half of the cycle involves the growth of the tooth and attachment to the jaw after eruption. Midway through the cycle (10–12 weeks), a new tooth is initiated. The second half of the cycle involves growth of the RT and resorption of the FT, until it is shed at ∼20 weeks (Fig. 4A).

Fig. 4.

Timeline of tooth development in I. iguana. The time from shed to shed is ∼20 weeks in the animals examined in this study. (A) Normal development and development with damage to the crown. After an FT is shed, the RT erupts within a few weeks. The RT continues to develop and attach to the bone for another ∼8 weeks. Once it is fully attached, a new tooth is initiated in the SL at approximately 10–12 weeks. This new tooth then grows and resorbs the FT until it is shed. If the crown is removed, the base of the tooth still needs to be resorbed before the RT can erupt. (B) Shuffled timeline in the newly treated animals, where teeth were extracted when the SL had differentiated into a cap-stage tooth, ∼6 weeks before the FT would normally shed (Week 14 versus Week 20). Initiation of the next tooth begins 10–12 weeks after extraction, 4–5 weeks earlier than would be expected with normal shedding (Weeks 10–11, asterisk). The total replacement cycle is still 20 weeks, however, has been shifted ahead 4–5 weeks.

The time between shedding events in this study, ∼20 weeks, is different from previous reports of tooth shedding intervals in the green iguana (Edmund 1969; Kline and Cullum 1984, 1985), which was reported as every 6 weeks in very young animals and 13 weeks in larger animals. There are several possibilities for this discrepancy. The first is that Kline and Cullum (1984, 1985) used a wax-bite technique to collect their data on a much finer scale than Edmund’s study, and so more replacement events could have been captured because sampling was more frequent. Second, the ages of the animals could affect tooth replacement rates, as rates are known to slow down with age (Kline and Cullum 1984; Erickson 1996; Brown et al. 2015), and the ages of Edmund’s animals were not known (although snout–vent lengths reported in the notes match those of 1- to 2-year-old iguanas). Third, the health of the animals, intraspecific variability, diet, photoperiod, or seasonality might affect tooth replacement rates, however little is known about the relationships between these factors (Cooper 1966). Given that the timing is consistent between Edmund’s three study animals, our new histological data on tooth development, and that R349 was examined bi-monthly post-extraction (Supplementary Fig. S4), we are confident in our interpretations of the patterning and timing data obtained in this study.

The longitudinal X-ray dataset analyzed here does not show evidence for an increase in tooth replacement rate after premature tooth extraction or damage. When the FT is extracted, if the next generation RT has begun to resorb the ankylosed portion of the FT, the RT is not affected by the premature removal of its predecessor and erupts normally (Fig. 4A). The third-generation tooth, which had not yet initiated at the time of surgery (given that only two teeth are present in one tooth family at a given time), also erupted as would be predicted. In the cases where teeth were incompletely removed and the base of the tooth was left attached to the jaw, that base would still need to be resorbed before the next tooth could erupt, thus maintaining normal, expected replacement timing (Fig. 4A). Therefore, breaking a tooth crown does not affect resorption or replacement, as suggested by Edmund (1960), and in more recent experiments in salmon (Huysseune et al. 2012). The time required for tooth resorption and growth of the RT is therefore an integral part of maintaining the regularity of the tooth replacement cycle.

These results differ from a similar experiment performed in mice, where a clipped incisor grows faster than expected following damage (An et al. 2018). The differences between reptiles and mammals could be due to the fact that mice incisors are ever-growing, and so a tooth does not need to be resorbed and an entirely new tooth does not need to develop following damage. Also, these differences could be attributed to the innervation and vascularization of the teeth, where the continual growth of the incisor in mice is due to a neurovascular bundle that serves a mesenchymal stem cell niche (Zhao et al. 2014). Iguana teeth are also innervated (Suzuki et al. 2007), and in fish innervation (Tuisku and Hildebrand 1994) and vascularization (Crucke and Huysseune 2015) are integral to tooth replacement. However, there is currently no data linking innervation and vascularization to tooth replacement and resorption in reptiles.

An examination of the molecular data shows that in general, the SL of the iguana is quiescent, with a lack of proliferative cells, until a new tooth is initiating. The presence of LRCs (potential stem cells) in the lingual side of the DL, dorsal to the SL, is similar to the patterns seen in other lizards: leopard geckos (Handrigan et al. 2010) and the pleurodont teeth of the bearded dragon (Salomies et al. 2019). However, in the leopard gecko, proportionately more cells are labeled in the SL after a 1-week pulse of BrdU than in the iguana and the bearded dragon. This difference might be due to the speed at which leopard geckos replace their teeth (each tooth is shed approximately every 30 days, Grieco and Richman 2018), so the rest period between initiation events is shorter. Conversely, in the alligator, LRCs are seen at all developmental stages in the DL bulge (possibly equivalent to the SL of squamates). When teeth were removed prematurely in the alligator, a boost in proliferation (an increase in TA cells) and more teeth in the initiation stage of growth were observed 1-week post extraction (Wu et al. 2013). In the iguana, there was no burst in proliferation 1 week after extraction, and proliferation did not occur until 12 weeks later (Fig. 3).

Nuclear β-Catenin can be detected in the SL after FT removal. This suggests the activation of canonical Wnt signaling. However, while nuclear localization is used as a marker for active signaling, we do not know if it is indicative of target gene transcription in the iguana. In alligator, leopard gecko, and snake teeth, analyses have shown that Wnt pathway activation plays an important role in the initiation of tooth development (Handrigan et al. 2010; Handrigan and Richman 2010b; Gaete and Tucker 2013; Tsai et al. 2016). Further studies are needed to know whether it really is an indicator of target gene transcription in the iguana and other reptiles.

For the new extraction experiments, only teeth that were firmly attached to the jaw with little to no resorption of the base from the RT were removed. Based on the 20-week shed cycle, teeth were removed ∼6 weeks ahead of the normal shedding time. Proliferation and initiation of a new tooth were observed 12 weeks after the extraction, which means that the initiation of the third-generation tooth occurred ∼4 weeks earlier than would be expected on a 20-week cycle (Fig. 4B). Therefore, it is possible that even though the immediate response of the dental tissues to tooth extraction was lacking when compared to the alligator, tooth replacement timing could increase in the iguana given the initiation of new teeth earlier than expected at that tooth position.

A consideration of the molecular control over tooth initiation must be considered in order to interpret the results of these experiments in the iguana. Several mechanisms have been proposed for the initiation of the alternating tooth replacement pattern observed in most polyphyodont animals (Edmund 1969; Osborn 1970, 1971; Westergaard and Ferguson 1987; Murray and Kulesa 1996; Fraser et al. 2006a; Huysseune and Witten 2006; Lamoureux et al. 2018; Gibert et al. 2019; Sadier et al. 2020). One theory suggests that signaling between neighboring teeth and a reaction-diffusion model control the spacing and timing and tooth initiation (Osborn 1971; Fraser et al. 2006b). However, it is unknown if these same signals control the patterning of teeth post-initiation, and if signaling between teeth in the same family is key in maintaining the timing of replacement rather than signaling between neighboring teeth.

In this experiment, we predicted that after removal of the FT, the RT would be able to erupt quickly given that it would no longer need to resorb the base of its predecessor. Indeed, it has been observed in studies of normal tooth eruption in crocodiles that eruption can be very rapid, >1–2 days after shedding of the FT (Finger et al. 2019). In the iguana, if the predecessor erupted more quickly than usual, then signaling along the DL could have been affected, and a new tooth would be initiated earlier than expected. Further experiments involving DL removal are needed to test whether signaling is controlled between adjacent teeth or within the tooth family by affecting the DL and SL.

Differences in the responses to premature tooth extraction between lizards and alligators could also be due to the type of tooth attachment style. The iguana, and most squamates, have a pleurodont attachment, where only the labial side of the tooth attaches directly to the jaw and there is no periodontal ligament (Bertin et al. 2018). The alligator has a thecodont attachment, where the tooth is implanted in a deep socket with a periodontal ligament. The process of replacement is different between these two types of attachment because in pleurodont animals the tooth needs to be detached from the bone by resorption before shedding the crown. In alligators, the tooth needs to be loosened from the attachment to the periodontal ligament. Removal of teeth in each animal is quite different and more invasive in the alligator, where soft tissue damage is higher due to the tearing of the ligament, which is very close to the dental epithelial bud. Damage to the soft tissue in the iguana is not as severe, although does happen, especially in the oral epithelium (Fig. 2A). Potentially, the increase in proliferation in the alligator and the increase in Wnt signaling in the alligator and iguana (Fig. 2C′) 1-week post extraction may be initially due to repair of the soft tissues before tooth development can begin again, as Wnts are often active in tissue repair (Zhao et al. 2009; Whyte et al. 2012).

Conclusions

Our study has revealed that the initiation and timing of tooth development in polyphyodont reptiles post-damage or extraction varies, specifically between iguanas and alligators. Damage to a tooth crown does not seem to affect the overall timing of tooth replacement; however, more severe damage such as extraction shifts the entire tooth replacement cycle ahead at that position by removing the need for resorption of the FT. Future studies will focus on how localized disruptions to the DL and RT removal affect the timing and patterning of replacement around the whole jaw, cellular and molecular changes, and if removing communication between neighboring teeth affects patterning. More longitudinal studies will determine if premature tooth extraction does in fact instigate faster tooth replacement around the entire jaw for a better understanding of tooth replacement dynamics in reptiles.

Data availability statement

The X-ray data examined in this study is available online through the University of Manitoba Dataverse:https://dataverse.lib.umanitoba.ca/dataverse/iguanaxrays (Brink 2020)

Animal ethics statement

The BrdU labeling and FT removal procedures were approved by the local institutional animal care and use committee at the University of Southern California.

From the symposium “Biology at the Cusp: Teeth as a Model Phenotype for Integrating Developmental Genomics, Biomechanics, and Ecology” presented at the annual meeting of the Society for Integrative and Comparative Biology, January 3–7, 2020 at Austin, Texas.