Healing of injured or degraded connective tissues requires migration, proliferation, and differentiation of local stromal cell populations. However, the low proliferative rate of cells in these tissues restricts the extent of healing. Cell transplantation is a potential approach to overcome intrinsic limitations of proliferation and differentiation in connective tissues. Cells of defined differentiation and proliferative potential can be expanded in vitro and then, using stable genetic markers, the contribution of the transplanted cells can be measured in vivo. The periodontium is a convenient and powerful model for studies of cell transplantation in complex connective tissues since cells of different phenotypes are derived from multipotent proliferating precursors (McCulloch and Bordin,1991) and are localized as islands of cells in the perivascular sites of the periodontal ligament (PL) (Gould et al.,1977; McCulloch,1985) and in the periosteal and endosteal spaces of the alveolar bone (Aukhil et al.,1990). The progeny of proliferating cells migrate to produce more differentiated cells that can synthesize collagen, bone, cementum, and the extracellular matrix of the PL, including other growth modulating molecules (Melcher,1988). To improve the understanding of cell migration, proliferation, and differentiation, we used combination of orthodontic tooth movement, thus disrupting the PL tissue homeostasis and implantation of PL or bone marrow cells to enhance tissue regeneration. Our results show that the tooth movement combined with cell transplantation promotes periodontal regeneration. The transplanted PL or bone marrow cells migrate to alveolar marrow spaces to home and proliferate. These cells then migrate to healing PL tissues to differentiate into soft (PL fibroblasts) and mineralized connective tissue cells (osteoblasts and cementoblasts).
Direct transplantation of multipotent precursor cells into the periodontium could provide a therapeutic approach for restoring periodontal tissues destroyed by periodontitis or trauma. To improve the understanding of cell migration, proliferation, and differentiation, we used a rodent model combining orthodontic tooth movement and transplantation of Lac-Z-positive murine-cultured periodontal ligament (PL) or femur-derived bone marrow precursor cells into a defined mandibular wound site, thus promoting tissue regeneration in wounded periodontium. Our results show that in orthodontically traumatized tissues, transplanted PL and bone marrow cells migrated systemically, contributing to the repopulation of sites with reduced cell/matrix density. The transplanted PL cells proliferated in adjacent alveolar bone marrow spaces, thus migrating to vascular tissues in the PL. The capillary walls in the PL serve as delivery sites for these cells and other marrow-derived hematopoietic cells, including monocytes. The transplanted marrow cells, extracted from femur of transgenic (TgR) mice exhibited similar behavior to those of transplanted PL cells, showing high proliferative activity in alveolar marrow as well as intensive repopulating capacity in wounded periodontium. On the other hand, the buccal skin fibroblasts failed to migrate and home effectively and thus the transplantation of these cells had no effect on periodontium regeneration. Based on these results, we conclude that the transplanted PL and bone marrow cells migrate systemically and following a cyclical process of growth and development and differentiate into PL fibroblasts, osteoblasts, and cementoblasts, thereby contributing to periodontal regeneration. © 2005 Wiley-Liss, Inc.
MATERIALS AND METHODS
Homozygous ROSA 26, 6- to 8-week-old C57BL/6J;B6 TgR male mice that express β-galactosidase (β-gal; Lac-Z reporter gene; Jackson Laboratories, Barharbor, ME) and β-gal+, green fluorescent protein-negative Sprague-Dawley male rat (Charles River) weighing 110–130 g were used. The animals received food and water ad libitum and were kept in a room with a 12/12-hr light/dark cycle. The rats were used in the present study because the anatomy of rat molars resembles human molars and, most importantly, rats have been shown to be useful in assessing biological response to tooth movement (Roberts and Chase,1981).
Primary cultures of PL cells, bone marrow cells, and buccal skin fibroblasts were established from Lac-Z-positive TgR mice. These cells were used as explants for recipient Sprague-Dawley rats. The mice were sacrificed by cervical dislocation and a block of mandibular tissue consisting of the mandibular first molar root, the PL, and a tiny stretch of the surrounding alveolar bone was obtained. PL cells that were collected with this method included precursor cells from the root-related and bone-related compartments of the PL as well as precursor cells located in the endosteal spaces of the alveolar bone. The bone marrow cells were collected from the femur. The femur was cleaned and extraneous tissues were removed. A section of femur inferior to the lesser trochanter and superior to the medial epicondyle was removed. Using a 1 ml syringe, a 26 G needle, and 5 ml of complete Dulbecco's modified Eagle's medium containing 10% bovine fetal serum, 0.25 μg/ml fungizone, 100 U/ml penicillin, and 100 μg/ml streptomycin, the marrow cells were flushed into a sterile tube. The cells were gently resuspended and isolated cells were cultured in 25 cm2 tissue culture flasks in a humidified incubator containing 5% CO2. The primary culture of buccal skin fibroblasts was established from a block of skin tissue from the cheek. In order to address the validity of the cell tracking method, we have for one of the controls used cultured green fluorescent protein-labeled murine embryonic stem cells (ES) derived from FVB Gc-Tg (GFPU 05 Nagy strain of transgenic mice; Jackson Laboratories). Cells grown in vitro for 4–5 subcultures were used for grafting into the recipient rats (Lekic et al.,2001).
In this study, a total of 60 rats were divided into various treatment and control groups as outlined in Table 1. Surgery was performed between 10:00 and 12:00 AM on the third day after the arrival of animals. Animals received orthodontic tooth movement (OTM) 1 day before cell transplantation by inserting separating elastic between the first and second mandibular molar (Fig. 1), thus causing profound disruption of periodontal tissue homeostasis (Lekic et al.,1997). Ex vivo expanded PL, buccal skin fibroblasts, and bone marrow cells that express tracking β-gal Lac-Z marker as well as ES fibroblasts that express green fluorescent protein were trypsinized, washed, and suspended in an Eppendorf tube either in sterile phosphate-buffered saline (PBS, pH 7.35) or sterile medium without bovine fetal serum. Using a dental probe, approximately 5 × 103 cells were transferred to the mandibular wound site under a stereomicroscope (Gould et al.,1977; Lekic et al.,2001). Animals were closely observed and kept warm for 1 hr after surgery and were sacrificed at 3, 6, and 12 hr following cell transplantation. The presence of transplanted cells was assessed in systemic circulation by obtaining samples of blood from the heart at 3, 6, and 12 hr after transplantation. Samples were cytospinned and enzymatically stained for β-gal in X-Gal buffer (5 mM potassium ferrocyanide, 5 mM potassium ferricyanide, 1 mM magnesium chloride) containing 1 mg/ml X-Gal for 3 hr at 37°C (Lekic et al.,2001). The animals sacrificed at 12 hr following topical xenografting showed transplanted cells in systemic circulation. In further transplantation studies, we examined two time points, 1 day and 3 days, since these time points correspond to periods of clot stabilization, early migration and proliferation, as well as early matrix formation in periodontal wound healing (Lekic et al.,2001). An hour prior to sacrifice, animals were injected intraperitoneally with 20 μCi/mmol of 3H-thymidine at a dosage of 1 μCi/g of body weight to assess the cell proliferation and migration (McCulloch and Melcher,1983).
|Group||Treatment||Number of animals|
|1 Day||3 Day|
|Control 1a||OTM + no transplanted cells||3||3|
|Control 1b||No OTM + cell transplantation||3||3|
|Control 2||No OTM + no transplanted cells||3||3|
|Control 3||OTM + GFP+ ES cells||3||3|
|Experimental 1||OTM + β-Gal+ PL cells||5||5|
|Experimental 2||OTM + β-Gal+ bone marrow cells||5||5|
|Experimental 3||OTM + β-Gal+ buccal skin fibroblasts||5||5|
Following the sacrifice of animal, mandible was removed and immediately fixed in periodate-lysine-paraformaldehyde for 24 hr at 4°C (Lekic et al.,1997). Demineralization was performed in 12.5% EDTA (pH 7.8) for 3 weeks and specimens were washed in PBS and incubated overnight with staining reagents for β-gal (X-Gal; Gibco). X-Gal-stained specimens were embedded with unpolymerized monomer for 4 days in a vacuum chamber (Immunobed; Polyscience) (Lekic et al.,2001). Frontal and sagittal sections, 5 μm in thickness, were prepared using JB-4 A Porter-Blum microtome. Serial sections were transferred to a water bath and placed on precleaned glass microscope slides. Slides were dried on a slide warmer and stored at 4°C.
The presence of apoptotic cells was determined with terminal deoxynucleotidyl transferase (TdT)-mediated dUTP nick end labeling (TUNEL) and confirmed by cysteinyl aspartate-specific protease-3 (caspase 3) assays (ApoTag Manual; Intergen) (Kainulainen et al.,2003). TUNEL-stained slides were prepared for immunohistochemistry (Chano et al.,2003).
Radioautographic slides containing mandibular molar tissue sections were dipped in full-strength Kodak NTB-2 emulsion, dried, and the air-dried slides were stored in light-tight dry boxes for 2 weeks at 4°C. After exposure, the slides were developed in Kodak D-19 developer and stained with hematoxylin and eosin through the emulsion (McCulloch and Melcher,1983). The cell was considered labeled if more than five silver grains overlaid its nucleus (P < 0.001). Background counts obtained from sections of control (unlabeled) animals and application of the Poisson distribution were used to derive this figure. The Poisson distribution was chosen for establishing counting thresholds because it fitted best with the raw grain count data.
In order to assess the disruption of tissue domains, the PL widths were calculated for coronal, middle, and apical segment of PL on tension and pressure sides. A BX41 Olympus microscope attached with a DP12 digital camera was used to analyze the stained and immune-labeled cells. The percentage of apoptotic and labeled cells in each landmark segment of the PL and the corresponding segment of alveolar bone (AB) were calculated on the tension and pressure sides for various time points using a grid ocular eye piece consisting of 100 squares, each measuring 50 × 50 μm. The junction of cellular and acellular cementum serves as the midpoint for the middle segment. Analysis was done for site-specific regeneration in the PL and the alveolar bone and within different regenerative zones of the mesiobuccal root of the mandibular first molar (Chano et al.,2003).
The selected slides were analyzed for immunohistochemical localization of osteopontin (OPN), bone sialoprotein (BSP), stromal cell marker (STRO-1), and ED-1 (antimacrophage) expressing cells. The mouse antirat OPN, mouse antirat BSP, and mouse antihuman stromal cell surface marker monoclonal antibodies (mAbs) were obtained from the Developmental Hybridoma Bank, University of Iowa. The mouse antirat macrophage mAb was obtained from Serotec (Ontario, Canada). Initially, the slides were blocked with avidin/biotin in blocking solution (Vector Laboratories, Ontario, Canada), then with 0.1% mouse normal serum diluted in 1% casein blocking solution and finally with 3% H2O2 in PBS. The primary antibodies were diluted in blocking solution. The OPN, BSP, and STRO-1 dilutions were 1:600, 1:400, and 1:200, respectively. The slides were incubated with primary antibodies overnight at 4°C in a moist chamber. The slides were treated with biotinylated horse-antimouse secondary antibody (dilution 1:200) and were incubated with avidin and biotin complex (Vector Laboratories). The color development reaction was obtained with DAB reagents (Vector Laboratories). Several controls were used during the staining procedure.
Tissue Domains and Cell Death
The average percentages of apoptotic cells per landmark site (coronal, middle, and apical) in the PL and the surrounding AB (control group 2; no tooth movement and no cell transplantation) were 14% and 20%, respectively. The PL width in the same group was 0.25 mm; however, in control group 1 (tooth movement and no cell transplantation), the PL width was greater on the tension side (0.27:0.18 mm; Fig. 2b). There were no significant changes in width or the number of apoptotic cells in the examined PL and AB segments of the control group 2.
The PL widths and the percentage of apoptotic cells in various segments of the PL and the corresponding AB segments showed a significant increase in the PL width in the apical segment of the tension side compared to the control group 2 without tooth movement (P < 0.05; Table 2). The percentages of apoptotic cells in the PL and AB of the treated groups were also significantly higher (P < 0.05) when compared to the control group 2 (data not shown).
|1 day experimental group||Distal (tension side) - PL||Mesial (pressure side) - PL|
|PL width (mm)||Percentage of apoptotic cells||PL width (mm)||Percentage of apoptotic cells|
The tension side in the 1-day experimental group showed the highest number of apoptotic cells (45–58% in PL; 28–32% in AB) compared to both the control group 2 (14% in PL; 20% in AB) and 3-day experimental groups (26–37% in PL; 23–33% in AB; Table 3). On the other hand, there were no significant changes in the percentages in apoptotic cells on the pressure side between the two time points except that the apical segment in the day 1 group showed a higher number of apoptosis compared to the corresponding segments in the day 3 group. In the middle and apical segment of the cellular cementum, approximately 20–27% of the cementoblasts showed apoptosis in the experimental groups. On the other hand, in controls, 15% and 19% of cementoblasts exhibited apoptosis in the middle and apical segments of cementum, respectively. In the experimental groups, apoptosis involved odontoblasts. Occasionally, clusters of odontoblasts were apoptotic, which is an uncommon finding since apoptosis often involves single cells. It is interesting to note that in the control group 1a, apoptotic cells were found throughout the PL.
|3 day experimental group||Distal (tension side) - PL||Mesial (pressure side) - PL|
|PL width (mm)||Percentage of apoptotic cells||PL width (mm)||Percentage of apoptotic cells|
Migration and Proliferation of Transplanted Cells
We found that most of the transplanted dermal cells were present only at the implant sites (Fig. 3a). However, the PL and bone marrow cells migrated shortly after transplantation becoming involved in the healing of wounded tissues (Figs. 2a and 3b). Indeed, the bone marrow or PL transplanted cells appear in systemic circulation 12 hr after topical placement of cells in mandibular wound site of animals undergoing tooth movement (Fig. 4). The alveolar marrow in transplanted animals undergoing tooth movement demonstrated a significant number of proliferating transplanted cells (25–45%; Fig. 5a). However, there was reduced cell proliferation in the bone marrow of animals undergoing tooth movement only (Fig. 5b). Notably, the same group of animals showed the presence of intensive cell proliferation throughout the wounded PL (day 3; Fig. 6a). By contrast, the PL cell proliferation was not increased in the tooth movement and PL cell transplant group (Fig. 6b). These observations clearly demonstrate that the transplanted precursor cells home in the alveolar marrow, where they proliferate and eventually migrate to wounded PL tissue. In support of this was the migratory and repopulation behavior of transplanted ES cells that expressed green fluorescent proteins (Fig. 7). It is important to note also that for the duration of our study, the xenografting did not cause any inflammatory or immunological response. This was confirmed by immunohistochemical analysis using mouse antirat macrophage antibody and histochemical examination of inflammatory cells including neutrophils (data not shown).
Many of the labeled transplanted cells were clustered in and around blood vessels in the PL and the alveolar bone. The internal elastic lamina of tunica intima of PL arteriole showed remodeling in transplanted animals. There were a few highly elongated spindle-shaped narrow cells, which were perhaps derived from bone marrow monocytes in the PL of treated animals (data not shown). The function and role of these cells remain unclear. The OPN-labeled cells were predominantly found around and within the blood vessels of the PL and vascular areas in the AB as well as periosteal and endosteal lining structures. In wounded periodontal tissues, the banded and segmented neutrophils were found to be BSP-positive (data not shown). On the other hand, BSP-labeled cells were detected in stromal and vascular tissues in the AB and the PL with a higher number of labeled cells found in AB. The STRO-1 (antihuman mesenchymal antibody) detected four structurally different types of cells. The cells detected by this antibody in the PL and the AB appear to have a typical mesenchymal cell shape. However, some of the STRO-1-positive cells in the PL were fibroblast-like spindle-shaped and some were circular-shaped blood progenitor cells (data not shown). When we examined the distribution patterns of repopulating cells in the PL following tooth movement, we found intense β-gal staining on the pressure side as well as near the root of the tooth (Fig. 2a). Similar observations were also made when GFP-labeled cells were transplanted (Fig. 7). The tension side of the PL had less β-gal and GFP staining compared to the pressure side. In the pressure side of the PL adjacent to the AB, there were a few (< 10) multinucleated, poorly defined, ruffled, bordered osteoclasts involved in bone resorption. On the other hand, the osteoblasts were far higher in number and were found in the tension side of the PL and AB, the endosteal, periosteal bone surfaces, as well as within blood vessels of the PL. The low number of osteoclasts may indicate that the number and the activity of osteoclasts are highly regulated by hormonal, humoral, and cellular factors, which are critical in bone remodeling under tooth movement.
Previous studies have shown that the regenerative capacity of the periodontal connective tissues is limited because major portion of the PL cell population does not renew (Davidson and McCulloch,1986) and the low proliferation rate of cells in the PL may restrict the extent of healing and contribute to the preponderance of reparative over regenerative processes. To overcome the loss of specialized cell types caused by disease, various therapeutic approaches have, with limited success, been developed to regulate and promote PL cell proliferation and differentiation. For example, the application of extracellular matrices, growth factors, and cell transplants (Pitaru et al.,1987; McCulloch and Bordin,1991; Rutherford et al.,1993). In this study, we postulated that in a tooth movement model with high amplitude of mechanical forces, followed by intensive cell death, the transplanted cells will home to their site of origin and replace lost specialized cell types and these cell types may contribute to the extracellular matrix. We understand that the Waldo method of tooth movement may cause tipping; nevertheless, this is a 4-day tooth movement and our pilot data have shown that during this time the separating elastic remains in place and minimal if any tooth tipping will occur. However, known limitations in using rats is the distance of molar teeth from the corner of the lip as well as the underdeveloped marginal and supra-alveolar fibers, for which the insertion of the separating elastic required surgical approach. Accordingly, this project was established to create a potential for the improved understanding of the origin as well as the fate of progenitor cell populations in mixed connective tissues. The main finding in this study is that transplanted PL and bone marrow cells systemically migrate to alveolar marrow to home and proliferate. After proliferation, these cells migrate to wounded periodontium to differentiate and repopulate apoptotic or injured cells induced by the orthodontic tooth movement. The capillaries and perivascular spaces in the PL and alveolar bone serve as the entrance sites into the wounded periodontal tissues. These cells appear in the capillaries as chain of cells, suggesting a ready recruitment of these cells by the wounded PL under tooth movement and mandibular wounding model. We tested four types of cells: β-gal+ PL cells, GFP+ ES cells, β-gal+ bone marrow-derived fibroblasts, and β-gal+ buccal skin fibroblasts. The PL and bone marrow-derived fibroblasts grown ex vivo for limited subcultures (4–5 passages) upon transplantation migrated and proliferated systemically to repopulate apoptotic cells. The presence of migratory and homing signaling mechanisms in wounded periodontal tissues had no effect on migration, growth, and proliferation of transplanted buccal skin fibroblast cells. We found that PL and bone marrow cells as well as ES cells were able to survive for a longer period compared to skin fibroblasts in vitro, indicating differential growth potentials between these cell types. We believe that the intracellular secretory molecules, the cellular microenvironmental niches, and the associated signaling mechanisms in wounded periodontium are critical determinants in defining the migratory and homing behavior and the fate of different cell types arising from transplanted multipotent precursor cells.
Studies of periodontal cell populations in steady-state conditions have limitations due to the relatively small number of proliferating cells (Gould et al.,1977). Alternatively, if the cell and tissue homeostasis is to be disrupted, then the precursor cells could be synchronously stimulated to proliferate, facilitating a more accurate assessment of the temporal expression of the repopulating cells. For this reason, we used tooth movement model to disrupt periodontal tissues and cell domains as well as to generate synchronized cohorts of proliferating and differentiating cells shortly after stimulation (Roberts and Chase,1981). Moreover, to assess the fate of the repopulating cells, we then transplanted labeled Lac-Z cells using the mandibular window wound model (Gould et al.,1977).
Our study suggests that transplanted periodontal or marrow cells exhibited similar migratory and proliferatory mechanism when repopulating wounded periodontal tissues. It is understood that both cell types have differentiating ability in vitro, and this may change in multiple subcultures (Guan et al.,2004). In fact, alveolar marrow cells can acquire PL characteristics in vitro as demonstrated by the immunohistochemistry using antirat SMA, OPN, and BSP monoclonal antibodies (Lekic et al.,2001). Guan et al. (2004) have reported similar observation stating that stem cells acquire PL characteristics in vitro. Pereira et al. (1995) have shown that mesenchymal cells expanded from marrow space in ex vivo act as precursors of bone and nonmineralized connective tissues.
Animal undergoing tooth movement and cell transplantation showed a limited number of proliferating transplanted cells in the PL compared to the animal receiving cell transplantation without tooth movement (control 1b, Table 1). We believe tooth movement provides a strong stimulus leading to a cascade of signaling mechanisms that result in multiple cellular activities including cellular proliferation in the alveolar marrow. Notably, transplanted PL or bone marrow cells undergoing several in vitro passages prior to their transplantation are further being influenced by the signaling mechanisms from the wounded tissues, thus exhibiting characteristics of more differentiated cells. For this reason, implanted PL or bone marrow cells showed lower proliferation rate in healing periodontal tissues when compared to nascent cell population. Similar studies have been reported showing the migration and proliferation of transplanted endothelial cells and subsequent tissue regeneration in diseased myocardium (Kawamoto et al.,2001), as well as replacement of dead host neurons with implanted fetal neurons (Isacson,1994). Collectively, the compartmentalized growth and differentiation of committed transplanted cells is a complex biological phenomenon. The functional role of transplanted cells relates to several factors: the cellular and genetic makeup of the transplanted cells and host's cellular and microenvironmental niches as well as the physiologic health of the host.
Based on our results, we concluded that the transplanted periodontal ligament and bone marrow precursor cells are of the same lineage and after transplantation they both follow the same route of systemic migration and proliferation in the alveolar bone marrow. Following a cyclical process of growth and development, these cells migrate to regenerate wounded periodontal tissues. It is important to point out that although transplanted cells replaced apoptotic cells of different phenotypes, differentiation and transdifferentiation of endogenous cell populations including epithelial cell remnants of Hertwig's sheath (Malassez's rests) require special consideration. We also found that transplantation of PL or bone marrow cells resulted in a number of GFP-positive alveolar marrow cells possibly due to signal transduction-based cell-cell communication. Overall, results from this study will improve our understanding of cell proliferation and migration in mixed connective tissues and hopefully lead to the development of more successful approaches for the treatment of periodontal diseases.
We acknowledge the work of Dr. Hanadi Nuseir in providing technical assistance and doing significant lab work for parts of the study.