|Home | About | Journals | Submit | Contact Us | Français|
The ability to use autologous dental progenitor cells (DPCs) to form organized periodontal tissues on titanium implants would be a significant improvement over current implant therapies. Based on prior experimental results, we hypothesized that rat periodontal ligament (PDL)-derived DPCs can be used to bioengineer PDL tissues on titanium implants in a novel, in vivo rat maxillary molar implant model. Analyses of recovered implants revealed organized PDL tissues surrounding titanium implant surfaces in PDL-cell-seeded, and not in unseeded control, implants. Rat PDL DPCs also exhibited differentiative potential characteristic of stem cells. These proof-of-principle findings suggest that PDL DPCs can organize periodontal tissues in the jaw, at the site of previously lost teeth, indicating that this method holds potential as an alternative approach to osseointegrated dental implants. Further refinement of this approach will facilitate the development of clinically relevant methods for autologous PDL regeneration on titanium implants in humans.
Strategies to improve the durability and function of titanium dental implants include: modified implant shape and design (Chen et al., 2005; Lee et al., 2005); altered surface topography to control cell behavior (Oates et al., 2007); nanostructured surface coatings (Catledge et al., 2002; Wang et al., 2005; Schliephake et al., 2006; Hedia, 2007); and the use of growth factors (Park et al., 2006). Although osseointegration is currently considered as the optimal implant/bone interface (Moradian-Oldak et al., 2006), naturally formed teeth are attached to surrounding alveolar bone via soft periodontal ligament (PDL) tissues. Since PDL-derived progenitor cells have been shown to self-renew, differentiate to multiple lineages, and function in periodontal tissue regeneration (Ivanovski et al., 2006; Kaneda et al., 2006; Nagatomo et al., 2006), their use in PDL tissue engineering has emerged as a promising approach for the treatment of periodontal disease. A critical requirement for successful PDL regeneration is the successful delivery of PDL progenitor cell populations capable of proliferating and differentiating in situ, in the oral environment.
A variety of scaffold materials have been used in PDL tissue engineering, revealing requirements for appropriate biocompatibility, biodegradability, cell adhesion, growth factor delivery, and mechanical stability (Abukawa et al., 2006; Moioli et al., 2007). Here we describe the use of Matrigel (BD Biosciences, Franklin Lakes, NJ, USA), a basement membrane matrix extensively used to study cell differentiation (Kleinman and Martin, 2005), to direct functional PDL tissue formation on titanium implants. We predict that the proposed methods will guide the development of clinically relevant therapies in humans.
Animal work was performed according to approved Tufts University IACUC protocols and National Institutes of Health guidelines. Maxillary first and second molars (M1 and M2) were extracted from 6- to 8-week-old female Lewis rats (Taconic, Germantown, NY, USA) under general anesthesia. Rats were administered palliative analgesics via subcutaneous injections twice/day and in the drinking water for 3 days post-implantation and were fed a soft diet.
Extracted molars and associated PDL tissues were pooled and rinsed in sterile Hanks’ balanced salt solution (HBSS). We obtained primary cell digests by incubating tooth roots at 37°C for 1 hr in 50 mL HBSS containing 0.67 mg/mL collagenase type II (Worthington Biochemical Corp., Lakewood, NJ, USA) and 0.3 mg/mL dispase (Roche, Basel, Switzerland). We obtained secondary cell digests by incubating primary digests a second time for 1.5 hrs at 37°C in Dulbecco’s modified Eagle’s medium (DMEM, Invitrogen, Carlsbad, CA, USA), 20% fetal bovine serum (FBS), containing 2.0 mg/mL collagenase type I (Worthington Biochemical Corp.), and 0.25% trypsin. Primary and secondary cell digests were washed, filtered through a 40-µm cell strainer (BD Falcon, Thermo Fisher Scientific, Pittsburgh, PA, USA), re-suspended in DMEM, 20% FBS, 100 U/mL penicillin, 100 µg/mL streptomycin, and cultured in humidified 5% CO2 at 37°C with media changes every 2 to 3 days.
Cultured rat PDL cell proliferation was monitored in triplicate with the PicoGreen dsDNA quantification kit (Invitrogen) and VICTOR3 fluorimeter (PerkinElmer, Boston, MA, USA). We analyzed the same samples for ALP activity, by mixing 80 µL of each with 100 µL p-nitrophenyl phosphate (Sigma-Aldrich, St. Louis, MO, USA), and 20 µL of 0.5 M 2-amino-2-methyl-1-propanol (Sigma-Aldrich) buffer, for 1 hr at 37°C. Spectroscopic measurements at 405 nm were compared with standard curve serial dilutions of 4-nitrophenol (Sigma-Aldrich).
Five independently harvested rat PDL cell preparations were plated at 2000 cells/T25 flask in DMEM, 20% FBS, cultured for 2 wks, fixed with 10% formalin, and stained with saturated methyl violet (Sigma-Aldrich). Colonies greater than 2 mm in diameter were counted and recorded.
The differentiation potentials of at least two independently prepared P3 PDL cell preparations were tested as previously described (Zhang et al., 2006). Briefly, adipogenic differentiation media contained 0.5 mM isobutyl-methylxanthine (IBMX), 1 µM dexamethasone, 10 µM insulin, 200 µM indomethacin, and 50 µg/mL of gentamicin. Neurogenic differentiation media contained 10 mM BME, 2% dimethyl sulfoxide (DMSO), and 200 mM butylated hydroxyanisole (BHA). Osteogenic media contained 5 mmol/L of KH2PO4, 10−8 M dexamethasone, 50 µg/mL of L-ascorbic acid, and 50 µg/mL gentamicin.
Rat PDL cell differentiation was assessed by IHC with primary antibodies anti-BSP (ab52128, Abcam, Cambridge, MA, USA), anti-DSPP (LF-153, Dr. Larry Fisher, NIH), anti-NeuN (MAB377, Millipore, Temecula, CA, USA), anti-STRO-1 (Invitrogen), and osteocalcin (OCN) (Abcam), with the Vectastain ABC staining kit (Vector Laboratories, Burlingame, CA, USA), and fast green counterstain. Negative control samples were assessed with secondary antibody alone.
RNA was isolated from triplicate, differentiated, and control PDL cell cultures with Trizol reagent (Invitrogen) and reverse-transcribed with the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA). qRT-PCR analyses were performed with the ABI Prism 7000 Sequence Detection System (Applied Biosystems), and Assays-on-Demand™ Gene Expression kits, for BSP (Rn01450118_m1), DSPP (Rn02132391_s1), and periostin (Rn01494630_g1), normalized to GAPDH (Rn99999916_s1) and analyzed with the ABI Prism 7000 Sequence Detection Systems version 1.0 software (Applied Biosystems).
Surface-coated custom implants (2.0 mm x 1.5 mm) were produced by Straumann (Basel, Switzerland) by a Sandblasting Large-grit Acid-etching (SLA) technique to provide macro roughness (Cochran et al., 2002). Following alveolar bone exposure, the implant site was prepared with a round bur (1.4 mm diameter) and custom-made drill (1.5 mm diameter) with mountable stop (2 mm). P3 cultured rat PDL cells were harvested, re-suspended in Matrigel (BD Sciences), and aliquoted (2 µL, 1 x 106 cells) into fabricated molded wells (1.7 mm diameter). An implant was placed into each well, and the Matrigel mix was hardened at 37°C for 1 hr. The resulting Matrigel coating was about 0.5 mm thick, and could withstand press-fitting. Implant beds were prepared to avoid friction during implantation, and the implants were quite stable. At least two independent cell preparations were used to generate duplicate ‘experimental cell plus Matrigel’ (C+M) and ‘Matrigel alone’ (M) implants. Uncoated titanium implants were used as positive ankylosis controls (Choi, 2000). Right side cell-seeded and left side negative control implants were placed at the healed M1 extraction site of each rat host.
Harvested rat palates were fixed in 10% buffered formalin ON. Our initial 7 hemi-mandibles with implants were sent for processing at the University of Minnesota Hard Tissue Research Laboratory (Donath and Breuner, 1982; Rohrer and Schubert, 1992). These samples were dehydrated, infiltrated with embedding resin (Technovit 7200 VLC, Kulzer, Wehrheim, Germany), polymerized at 450 nm, cut to 150 µm (EXAKT Technologies, Oklahoma City, OK, USA), polished to 35-50 µm with sandpaper discs (800-2400 grit, EXAKT micro-grinding system), final-polished with 0.3 µm alumina polishing paste, and stained with Stevenel’s blue and Van Gieson’s picro fuchsin, after epifluorescent and histomorphometric analyses. We chose to process the remaining 14 non-decalcified hemi-maxilla implants at the University of Bern. These samples were embedded in methylmethacrylate, sectioned longitudinally into ~400 µm ground sections by means of a slow-speed diamond saw (Varicut® VC-50, Leco, Munich, Germany), polished (grain size 4000) to a final thickness of ~150 µm (Knuth-Rotor-3, Struers, Rødovre/Copenhagen, Denmark), and surface-stained with toluidine blue and basic fuchsin (Schenk et al., 1984).
Implant sections were analyzed at the University of Bern with Leica stereolupe M8® and Dialux® EB light microscopes (Leica, Glattbrugg, Switzerland), 60-mm objective (Nikkor, Nikon), and Nikon DN 100® digital camera (Nikon, Egg, Switzerland). Implant osseointegration was evaluated directly at 160X with an integrative eyepiece for determination of percentage bone-to-implant contact (%BIC). Replicate blinded morphometric analyses were performed for reproducibility. Samples also were evaluated for PDL tissue and blood vessel formation/mm2.
Cultured P3 rat PDL cells continued to proliferate over 33 days (Fig. 1A). ALP activity in identical triplicate samples showed increased activity to day 20 and then declined to day 30 (Fig. 1B). Rat PDL DPCs harvested with either of 2 enzymatic digestion methods (primary and secondary digests) exhibited similar CFUs, ranging in diameter from 0.5 to 3.0 mm (Fig. 1C). Colonies larger than 2 mm were scored as positive. A statistically significant decrease in CFU formation was observed in later-passaged cells, suggesting that non-DPCs overgrew DPCs over time.
Adipogenic-induced P3 rat PDL cells exhibited oil red O-positive lipid-rich vacuoles, while control cells were negative (Figs. 1D and and1G,1G, respectively). Mineralized, Alizarin red (AR)-positive mineralized tissues were observed in osteoinduced cultures, while control cell cultures were negative (Figs. 1E and and1H,1H, respectively). Neurogenic-induced rat PDL cells exhibited neuronal-like cell bodies and cytoplasmic extensions, while controls did not (Figs. 1F and and1I,1I, respectively). Quantitative analyses of extracted AR stain revealed ~3X higher mineralization in osteoinduced cultures relative to control cultures (Fig. 1J). qRT-PCR revealed up-regulated BSP, DSPP, and periostin expression in osteogenic-induced PDL cell cultures, while periostin was down-regulated in adipogenic and neurogenic-induced cells (Fig. 1K). Immunohistochemical (IHC) analysis revealed ~10% STRO-1 positive (Gronthos et al., 2006) cultured rat PDL DPCs (Figs. 2A vs. .2E),2E), positive NeuN expression in neuronal-induced cultures (Figs. 2B vs. .2F),2F), and BSP and DSPP expression in osteoinduced PDL cells (Figs. 2C, ,2,2, vs. .2G,2G, ,2H2H).
Titanium implants were placed as described in Methods, and as shown in Figure 3. Out of 21 experimental and control implants, 2 exhibited implant site inflammation and were excluded from the study. Of the 8 Matrigel-coated non-cell-seeded implants examined, only 1 appeared osseointegrated, while the remaining 7 were surrounded by well-vascularized granulation tissue (data not shown), consistent with published reports for Matrigel (Cronin et al., 2004). As anticipated, none of the 9 Matrigel-coated cell-seeded implants was osseointegrated, but rather exhibited poorly vascularized, organized PDL-like tissue formation (Nanci, 2003) (Figs. 4A, ,4B).4B). Titanium-alone implants exhibited granulation tissue at the bone-implant interface at 8 wks and appeared osseointegrated at 18 wks (Fig. 4C). One out of 9 (10%) cell-seeded Matrigel-coated implants exhibited a thin layer of cementum-like tissue at the implant-PDL tissue interface, and poorly vascularized PDL tissues, with collagen fibers oriented perpendicular to the cementum surface (Fig. 4D), closely resembling naturally formed PDL and Sharpey’s fibers. In the remaining 8 experimental implants which lacked cementum, collagen fibers were oriented parallel to the implant surface (data not shown).
Since the embedding processes required to section titanium are not compatible with IHC analyses, we used IHC to examine titanium-free (M+C) and (M) implants. At 8 wks, no obvious bone formation was observed in either the (M) or (M+C) samples, although polarized light microscopy revealed organized collagen fibers in (M+C), but not in (M) implants (data not shown). At 12 and 18 wks, both (M) and (M+C) titanium-free implants exhibited newly formed bone and disorganized collagen fibers (Figs. 4E, ,4F).4F). Osteocalcin (OC) was weakly expressed in 8-week (M) and (M+C) samples (data not shown) (Ogawa et al., 2004). The 12-week (M)-alone implants showed strong OC expression (Fig. 4G), while, in contrast, only weak OC expression was observed in (M+C) implants (Fig. 4H), consistent with bioengineered PDL tissue formation.
In 1982, Nyman et al. demonstrated that PDL cells could be used to re-establish connective tissue attachment to teeth. Additional evidence for PDL tissue regeneration includes cementum deposition and collagen fiber attachment to dental implants (Buser et al., 1990; Warrer et al., 1993; Takata et al., 1995; Choi, 2000), and the formation of organized cementum, PDL, and bone on hollowed titanium implant surfaces (Parlar et al., 2005). These reports suggest new opportunities for implant dentistry, including the potential for autologous PDL tissue regeneration in humans. Here we report the utilization of autologous rat PDL cells derived from molar tooth root surfaces to regenerate PDL tissues on titanium implants grown in the jaw. Elegant studies have provided valuable information on PDL-derived DPC self-renewal, clonogenicity, differentiation, and capacity to re-establish connective tissue attachment between cementum and surrounding alveolar bone (Chen et al., 2006; Gronthos et al., 2006; Ivanovski et al., 2006; Kaneda et al., 2006).
Our results show that Matrigel, a three-dimensional biomatrix scaffold rich in essential ECM components (Kleinman and Martin, 2005), facilitates organized rat PDL regeneration at the titanium-implant-alveolar-bone interface, in this animal model. Although bioengineered cementum-like tissue was observed in only 10% of the PDL-cell-seeded experimental implants, associated collagen fibers appeared oriented perpendicular to the implant surface in this implant, as observed in naturally formed PDL tissues (Wolman and Kasten, 1986; Carter and Sloan, 1994; Komatsu et al., 1998). This observation suggests that proper collagen fiber alignment may facilitate subsequent cementum formation. Although 2/21 implants exhibited peri-implant infection, we anticipate that soft-tissue wound healing in humans will be much more predictable, because of effective oral hygiene, anti-inflammatory control, and implant site protection.
Our results show that PDL-derived DPCs could potentially be used to regenerate autologous PDL tissues on titanium implants grown in the jaw. Cultured PDL-derived DPCs exhibited high proliferation rates, clonogenicity, and adipose, bone, and neural cell differentiation. Bioengineered periodontal tissues formed cementum-like tissue on the titanium implant surface, and PDL tissue with Sharpey’s fibers inserted perpendicular to the implant. We propose that this study suggests the potential to replace missing teeth in humans with dental implants augmented with autologous cell-derived bioengineered periodontal tissues. We are currently working to refine these techniques using human PDL-derived DPCs, and alternative scaffold materials and designs, to eventually achieve reliable and effective, clinically relevant periodontal tissue regeneration in humans. In particular, methods to ensure the formation of PDL collagen fibers oriented perpendicular to the implant might improve cementum formation on the implant surface. Future consideration of inductive signals for PDL regeneration may improve PDL tissue regeneration on titanium implants, while maintaining soft- and hard-tissue interfaces.
This work was supported by the Center for Integrated Medicine and Innovative Technology (CIMIT), International Team for Implantology (ITI) Foundation, and NIH/NIDCR/NIBIB DE016132 (PCY). This study also was supported by award number K12GM074869 (TEACRS) from the National Institute of General Medical Sciences. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute of General Medical Sciences or the National Institutes of Health.