Bioengineered Teeth from Cultured Rat Tooth Bud Cells


Bioengineered Teeth from Cultured Rat Tooth Bud Cells

M.T. Duailibi4, S.E. Duailibi4, C.S. Young2, J.D. Bartlett2, J.P. Vacanti3, and P.C. Yelick2,*

1 University Federal of São Paulo, Department of Otorhinolaryngology and Human Communication Disorders, São Paolo, Brazil;
2 Department of Cytokine Biology, The Forsyth Institute, 140 The Fenway, and Department of Oral and Developmental Biology, Harvard Medical School, Boston, MA 02115, USA; and
3 Department of Surgery, Massachusetts General Hospital and Harvard Medical School, Boston, MA, USA;

* corresponding author, [email protected]

The recent bioengineering of complex tooth structures from pig tooth bud tissues suggests the potential for the regeneration of mammalian dental tissues. We have improved tooth bioengineering methods by comparing the utility of cultured rat tooth bud cells obtained from three- to seven-day post-natal (dpn) rats for tooth-tissue-engineering applications. Cell-seeded biodegradable scaffolds were grown in the omenta of adult rat hosts for 12 wks, then harvested. Analyses of 12-week implant tissues demonstrated that dissociated 4-dpn rat tooth bud cells seeded for 1 hr onto PGA or PLGA scaffolds generated bioengineered tooth tissues most reliably. We conclude that tooth-tissue-engineering methods can be used to generate both pig and rat tooth tissues. Furthermore, our ability to bioengineer tooth structures from cultured tooth bud cells suggests that dental epithelial and mesenchymal stem cells can be maintained in vitro for at least 6 days.

KEY WORDS: tooth tissue engineering • dental stem cells

Source: J Dent Res 83(7): 523-528, 2004.



The incidence of children born with missing primary and/or adult teeth is significant (Nunn et al., 2003), and tooth loss in aged populations is also a prevalent health problem. Current replacement tooth methods use synthetic materials that can elicit an immune induced host rejection response. The ability to generate biological tooth substitutes from autologous human tissues would be a valuable clinical tool. Tissue engineering—a relatively new science integrating developmental, molecular/cellular biology, and genetics with the field of engineering—holds promise in this area (Langer and Vacanti, 1993; Sittinger et al., 1996; Choi and Vacanti, 1997; Bohl et al., 1998; Kim and Vacanti, 1999; Mooney and Mikos, 1999; Stock and Vacanti, 2001; Vacanti et al., 2001). In the fields of maxillofacial surgery and periodontics, tissue engineering has been used to generate alveolar bone (Abukawa et al., 2003) and to regenerate oral tissues lost to cancer, decay, and periodontitis (Lynch et al., 1999). Cultured human dental pulp and gingival fibroblasts adhere to biodegradable scaffolds, and proliferate and differentiate in vitro (Mooney et al., 1996; Murphy and Mooney, 1999) and in vivo (Buurma et al., 1999). More recently, we have reported the successful bioengineering of whole tooth crowns composed of accurately formed enamel, dentin, and pulp tissues (Young et al., 2002).

The objective of this study was to improve upon our tooth-tissue-engineering methods by optimizing the age of tooth bud cells. In addition, we further define the progenitor cell populations giving rise to bioengineered tooth structures by exclusively using single-cell suspensions of rat tooth bud cells that were first cultured in vitro for 6 days. Finally, we compared the use of PGA and PLGA scaffold materials. The results of this study show that, as previously demonstrated for pig tooth buds, rat tooth bud cells can be used to bioengineer complex tooth crowns, suggesting a common application for mammalian tooth tissue engineering. Furthermore, our demonstrated ability to use 4-dpn tooth bud cells that were first cultured in vitro for 6 days suggests that epithelial and mesenchymal dental stem cells (DSCs) giving rise to bioengineered tooth structures can be maintained in culture. The results of this study significantly advance current tooth-tissue-engineering efforts by demonstrating a general application to mammals, and suggest a potential means to propagate and expand DSCs in culture.

Materials and Methods


Animal Husbandry
Lewis rats were maintained, and surgeries were performed, in the Forsyth Institute Animal Facility, following National Institutes of Health (NIH) and ALAC regulations, IACUC protocol #01-009, and animal assurance #A3051-1.

Polymer Scaffold Fabrication
Rectangular scaffolds (1 x 5 x 5 mm) were fabricated from polyglycolic acid (PGA) and poly co-glycolide copolymer (PLGA) as previously described (Young et al., 2002). Briefly, PGA fiber mesh containing 3% w/w poly(L-lactic acid) was packed into molds in chloroform, lyophilized for 48 hrs, and sanitized with 75% ethanol. We generated PLGA tooth scaffolds by packing polyvinylsiloxane molds half full with NaCl crystals, filling the remaining space with a 5% w/w, 85:15 molar ratio PLGA solution in chloroform, lyophilizing for 48 hrs, leaching the scaffolds in distilled water for 24 hrs, and sanitizing in 75% ethanol.

Isolation, Culturing, and Seeding of Rat Tooth Bud Cells
Molar tooth buds were isolated from 3- to 7-dpn Lewis rat pups and minced into balanced salt solution (HBSS, Gibco BRL, Gaithersburg, MD, USA). Tooth bud tissues were digested with type I collagenase (0.66 mg/mL, Sigma-Aldrich, St. Louis, MO, USA) and Dispase I (0.33 mg/mL; Boehringer Mannheim, Indianapolis, IN, USA), dissociated by trituration, and washed 5x in 50% Dulbecco’s modified Eagle medium (DMEM, Gibco BRL, Gaithersburg, MD, USA) containing 10% fetal bovine serum (FBS), 5 mL Glutamax, 50 units/mL penicillin, 50 µg/mL streptomycin, 2.5 µg/mL ascorbic acid, and 50% F12 medium (Sigma-Aldrich Corp, St. Louis, MO, USA). Single-cell suspensions were generated by filtration through a Falcon 40-micron cell strainer, typically yielding 2.4 x 105 cells/tooth bud. Cells were re-suspended in DMEM/F12, plated into 75-cm2 (T75) culture flasks (Costar, Cambridge, MA, USA) at 2.5 x 105 cells/mL, and grown in 5% CO2 at 37°C until the cells reached confluence at 6 days. Cells were harvested by trypsinization (0.25% trypsin/EDTA; Gibco-Invitrogen Corp., Tulsa, OK, USA) for 10 min at 37°C, washed twice with the same medium, recounted, split into equal portions, and statically seeded onto PGA and PLGA scaffolds for 1 hr prior to implantation into the omenta of syngeneic Lewis rat hosts.

Immunohistochemical analysis of cytokeratin expression in cultured epithelial tooth bud cells was performed with the use of the monoclonal pan-cytokeratin antibody PCK-26 (Sigma-Aldrich, St. Louis, MO, USA), according to the manufacturer’s recommended protocol.

Omental Implant Procedure
Adult Lewis rats (Charles River Laboratories, Wilmington, MA, USA), aged 6–12 mos, were used as hosts for tooth tissue implants. Omental surgeries were performed as previously described (Young et al., 2002).

Analyses of Implant Tissues
Radiographic analyses were performed with the use of a Hewlett-Packard Faxitron (Model 43855 TO-2) and Kodak (Rochester, NY, USA) high-speed SO-253 holographic film at 40 Kv and 3 mA for 30 min at a focal distance of 40 cm. After visual and radiographic inspections, implants were fixed in 5% formalin for 24 hrs, decalcified, embedded in paraffin, sectioned at 6-micron intervals, and stained with hematoxylin and eosin (H&E) or Goldner’s trichrome. Immunohistochemical analyses were performed using a polyclonal amelogenin antibody as previously described (Young et al., 2002). Sectioned and stained specimens were examined with the use of a Leica DMRE compound microscope and digital Zeis Axiocam digital camera (Stuttgart, Germany).



Characterization of Cultured Rat Tooth Bud Cells
To determine the age at which rat tooth buds were optimal for tooth tissue bioengineering, we prepared cultured dental cell populations from 3- to 7-dpn rat tooth buds. At least 3 experiments with a minimum of 6 rat pups (48 molar tooth buds) were performed at each developmental stage. After 6 days in culture, the dental cells appeared to be heterogeneous, consisting of fibrous, mesenchymal-like cells and clusters of smaller, epithelial-like cells (Figs. 1A, 1B). Cultured 5-, 6-, and 7-dpn tooth bud cells appeared to be dying at 6 days, exhibiting numerous floating cells and low total cell counts. In contrast, 3- and 4-dpn tooth bud cell cultures appeared healthy and exhibited average cell count/tooth bud of (2.0 x 105) and (1.5 x 105), respectively. Since 4-dpn tooth bud cells appeared to be proliferating in culture while 3-dpn tooth bud cells did not, 4-dpn rat molar tooth bud cells were selected for use in all subsequent tooth-tissue-engineering experiments.

We used immunohistological analysis of cytokeratin expression to identify epithelial cells in mixed epithelial and mesenchymal dental cell cultures. Brightly fluorescing cytokeratin-positive dental epithelial cells were readily identifiable under UV illumination (Fig. 1D, white arrows), while dental mesenchymal cells exhibited only background fluorescence. Positive control oral epithelium exhibited distinct cytokeratin immunoreactivity (Fig. 1C). Six-day cultured dental cells were harvested and seeded onto either PGA or PLGA scaffolds for 1 hr at 37°C in a humidified, 5% CO2 environment. Light microscopic analysis of cell-seeded PGA and PLGA scaffolds revealed cells attached to both polymer scaffolds (Figs. 1E, 1F, respectively).

Experimental and Control Implant Groups
Control groups C1-C3 consisted of: (C1) 7 non-dissociated 4-dpn molar tooth buds implanted as positive controls; (C2) 5 unseeded PGA scaffolds; and (C3) 5 unseeded PLGA scaffolds. Experimental groups E1 and E2 consisted of: (E1) 8 PGA scaffold implants seeded for 1 hr; and (E2) 8 PLGA scaffold implants seeded for 1 hr. Control and experimental implants were grown in the omenta of syngeneic adult rat hosts for 12 wks, as determined empirically by the detection of distinctly radio-opaque tissues in dental-cell-seeded scaffold implants.

Visual and Radiographic Analyses of Excised Implants
At 12 wks, experimental and control implants were excised and analyzed. By visual inspection, the implants appeared similar in color, size, and shape. Numerous experimental implants exhibited mineralized tissues protruding from the implant (Figs. 2A, 2B). Radiographic analyses of experimental implants revealed the presence of highly mineralized tissues (Figs. 3A', 3B'). Negative control, unseeded scaffold groups C2 and C3 contained no radio-opaque tissue (data not shown). A total of 7 out of 8 (88%) PGA and 4 out of 8 (50%) PLGA implants contained radio-opaque tissues.

Histological Analysis of 12-week Implants
We performed histological analyses to determine the cellular organization of mineralized implant tissues. All of the C1 control implants developed into accurately formed rat molar teeth containing identifiable dentin, enamel, and pulp (Figs. 3A, 3A'), although cementum and periodontal ligament tissues were not definitively identifiable in these implants at 12 wks. Histological analyses of PGA and PLGA scaffold cell-seeded implants demonstrated the presence of dentin, enamel, and pulp tissues (Figs. 3B, 3B', and 3C, 3C', respectively). Hertwig’s epithelial root sheath structures formed on both types of scaffold material. Infiltrating lymphocytes were evident in some of the sectioned implants (Fig. 3C', arrow).

The mineralized tissues of control and experimental implant groups were examined with the use of Goldner’s stain (Bancroft and Gamble, 2002), which stains dentin and bone blue, newly formed enamel matrix red, and mature enamel matrix gray (Z. Skobe, personal communication). Intact tooth bud control implants exhibited blue-stained dentin, red-stained newly formed enamel, and gray-stained mature demineralized enamel (Figs. 3D, 3D'). Similarly, tooth tissues bioengineered on both PGA and PLGA scaffolds exhibited blue-stained dentin, and gray-stained mature enamel (Figs. 3E, 3E', and 3F, 3F', respectively). Tooth tissues generated on PGA scaffolds generally exhibited more mature enamel that stained gray with Goldner’s (Fig. 3E'), while PLGA cell-seeded scaffolds generated both immature and mature enamel that stained reddish to gray (Fig. 3F).

Immunohistochemical Analysis of Bioengineered Rat Tooth Tissues
We used immunohistochemical analysis to examine the expression of amelogenin in bioengineered enamel. Control intact tooth bud implants exhibited positive amelogenin expression in ameloblasts and in demineralized enamel (Fig. 4A, arrows), while pre-immune control tissues were negative (Fig. 4A'). Bioengineered enamel grown on both PGA and PLGA scaffolds exhibited positive staining for amelogenin (Figs. 4B, 4B', and 4C, 4C', respectively).



Cultured Rat Molar Tooth Bud Cells Can be Used to Bioengineer Highly Ordered Tooth Structures in 12 Wks.
Our tooth-tissue-engineering approach is based on previous methods used to bioengineer neonatal intestine and mineralized tissues (Choi et al., 1997, 1998; Hutmacher et al., 2001a,b; Kaigler and Mooney, 2001; Ma and Choi, 2001). As previously demonstrated for pig tooth bud cells (Young et al., 2002), here we demonstrate that rat molar tooth bud cells, obtained from 4-dpn rats, can be used to bioengineer mature tooth crowns that appear highly characteristic of naturally formed teeth. Comparison of 3-, 4-, 5-, 6-, and 7-dpn rat molar tooth bud cells demonstrated that 4-dpn tooth bud cell cultures were optimal for tooth-tissue engineering. At 4 dpn, naturally forming rat molar teeth are cap-stage, with the most mature M1 molar just beginning to exhibit cytodifferentiation (Kollar and Baird, 1969). Isolated 4-dpn rat molar tooth bud cells exhibited the highest cell yield/tooth bud and viability after 6 days in culture. In addition, we have further refined previous tissue-engineering methods to involve the exclusive use of cultured tooth bud cells in this study. The exclusive use of single-cell suspensions of cultured tooth bud cells eliminates the possibility that bioengineered tooth structures arose from incompletely dissociated and/or differentiated tooth bud tissues. The fact that cultured tooth bud cells can give rise to bioengineered tooth structures suggests that the progenitor cells giving rise to these structures can be maintained in culture. Recently, characterization of mesenchymal DSCs from post-natal pulp (Gronthos et al., 2000) and from deciduous tooth pulp (Miura et al., 2003) has been reported, and the epithelial DSC niche has been recently described (Harada et al., 1999), although the epithelial DSCs themselves remain largely uncharacterized. Our demonstrated ability to use cultured tooth bud cells to bioengineer mature tooth crowns containing both dentin and enamel suggests that both epithelial and mesenchymal DSCs can be maintained in culture for at least 6 days.

Comparison of PGA and PLGA Scaffold Materials for Tooth-tissue Engineering
Our results demonstrate that PGA and PLGA scaffold materials are similar in their ability to support the growth of highly ordered dental tissues. These results are consistent with those from other reports where both of these scaffolds were used in the tissue engineering of maxillofacial tissues (Mooney et al., 1996; Buurma et al., 1999). In this report, we have not been able to examine host vs. donor contributions to bioengineered tooth structures, due to the fact that both male and female tooth buds were used in these experiments. Studies are currently being performed which use male tooth bud cells exclusively to seed scaffolds implanted into female hosts. These implants will be analyzed for the presence of Y-chromosome-positive bioengineered tooth tissues.

Comparison of Bioengineered Rat and Pig Teeth
Our previous results demonstrated that bioengineered pig tooth crowns containing dentin, pulp, and enamel formed in 25 to 30 wks. In contrast, rat tooth crowns formed in just 12 wks. The fact that bioengineered rat teeth developed more quickly than bioengineered pig teeth may reflect the natural growth patterns of these distinct mammalian teeth. Pig third molar teeth require approximately 80 wks to erupt (Bivin and McClure, 1976), while rat first molar teeth erupt after approximately 7 wks. Our results suggest that mammalian tooth bud cells may retain a cell-autonomous developmental program, even when dissociated into single-cell suspensions and grown in culture.

In summary, the results presented here support the following conclusions. First, we demonstrate the successful bioengineering of mature tooth structures from single-cell suspensions of cultured rat tooth bud cells. We demonstrate that 4-dpn rat tooth buds are optimal for this approach, and that bioengineered rat teeth develop reliably in 12 wks. We also demonstrate that PGA and PLGA scaffolds support the growth of mature tooth structures. We also note that implantation of Lewis rat tooth bud cells into syngeneic adult Lewis rat hosts demonstrates the successful use of autografts for tooth-tissue engineering. Future studies now need to address the underlying mechanisms responsible for directing the growth of these tooth structures.



The authors acknowledge the support and advice of Dr. Paolo Augusto de Lima Pontes, Full Professor of Otolaryngology, University Federal of São Paulo (UNIFESP), Brazil, and CAPES for support of MTD and SED, and for support by the Harvard School of D