Bisphosphonate modulates proliferation and ... - Wiley Online Library

0 downloads 0 Views 696KB Size Report
Periodontal Ligament Cells During Wound Healing. P. LEKIC,1* I. RUBBINO,2 F. KRASNOSHTEIN,1 S. CHEIFETZ,1 C.A.G. MCCULLOCH,1. AND H.
THE ANATOMICAL RECORD 247:329–340 (1997)

Bisphosphonate Modulates Proliferation and Differentiation of Rat Periodontal Ligament Cells During Wound Healing P. LEKIC,1* I. RUBBINO,2 F. KRASNOSHTEIN,1 S. CHEIFETZ,1 C.A.G. MCCULLOCH,1 AND H. TENENBAUM1,3 1MRC Group in Periodontal Physiology, Faculty of Dentistry, University of Toronto 2Samuel Lunenfeld Research Institute of Mount Sinai Hospital, Toronto, Ontario, Canada; 3Clinica Odontostomatologica, Universita di Firenze, Firenze, Italy

ABSTRACT Background: Periodontal ligament (PL) width is precisely maintained throughout the lifetime of adult mammals, but the biological mechanisms that regulate the spatial locations of the cell populations for bone, cementum, and PL are unknown. Methods: As bisphosphonates induce ankylosis in mouse molar teeth, we used ethane-1-hydroxy-1, 1-bisphosphonate—(HEBP, Etidronate; Didronel) in combination with a periodontal window wound model to identify those cell populations involved in the regulation of PL width during the reformation of cellular domains after wounding. Four groups of Wistar rats were wounded by drilling through the alveolar bone and extirpation of the PL. Rats were administered HEBP for 1 week and then sacrificed or allowed to recover for an additional week prior to sacrifice. Control rats were sacrificed after 1 or 2 weeks. One hour prior to sacrifice, rats were injected with 3H-thymidine to label proliferating cells. Tissue sections were immunohistochemically stained for osteopontin (OPN) or bone sialoprotein (BSP) or were prepared for in situ hybridization (BSP) to identify extra- and intracellular expression of these non-collagenous bone proteins associated with periodontal healing. Results: HEBP treatment for 1 week induced a twofold increase in the thickness of the alveolar bone matrix in which weak immuno-staining for OPN and BSP mRNA signal was seen. During the recovery phase the increased bone width was reduced but was still considerably thicker than in control (P F 0.001). OPN staining as well as the BSP mRNA signal were much more intense than at 1 week. HEBP induced a G 40% reduction of PL width which returned to normal dimensions following the recovery phase. HEBP also modulated PL cell proliferation and differentiation: PL cell counts and labelling indices were reduced fivefold after 1 week of HEBP but returned to control values after the recovery phase. In controls, PL cells did not express OPN and BSP, but after HEBP treatment, and particularly after the recovery phase, PL cells expressed both of these markers intensely. In contrast, gingival and pulp connective tissues that were contiguous with the PL were not stained for OPN and did not express BSP mRNA after HEBP treatment. Conclusions: While wounding induced transient increases of proliferation which were followed by repopulation of the extirpated tissue, the effects of HEBP on cell differentiation were independent of wounding. HEBP modulates the differentiation of PL cells and recruits cells that contribute to alveolar bone formation and loss of PL width homeostasis. Conceivably, bisphosphonates could be used therapeutically to selectively alter the differentiation of PL cells and promote the formation of alveolar bone and cementum. Anat. Rec. 247:329–340, 1997. r 1997 Wiley-Liss, Inc. Received 6 March 1996; accepted 25 September 1996. *Correspondence to: Dr. P. Lekic, Faculty of Dentistry, Room 344E, University of Manitoba, 780 Bannatyne Avenue, Winnipeg, Manitoba, Canada R3E 0W2. Contract grant sponsor: MRC of Canada Group Grant.

r 1997 WILEY-LISS, INC.

330

P. LEKIC ET AL.

Key words: periodontal ligament; fibroblasts; osteoblasts; bisphosphonate; alveolar bone The periodontal ligament (PL) is a complex, vascular, and highly cellular soft connective tissue that is interposed between two mineralized connective tissues, cementum, and bone. This tissue is important in the maintenance of tooth position and in the distribution of masticatory forces to the alveolar bone. The PL contains a mixture of cell populations including progenitors for cementoblasts, osteoblasts and fibroblasts. These cells are essential for physiological remodelling and for the healing of periodontal wounds (Melcher, 1976). The width of the PL is maintained throughout the lifetime of mammals at remarkably constant dimensions in teeth of limited eruption, indicating that there are biological mechanisms which tightly regulate the metabolism and spatial locations of cell populations involved in the formation of bone, cementum, and PL. For example, in spite of physiological tooth drift and the periodic application of high amplitude physical forces during mastication, the osteogenic activity of cell populations on the border of the PL is inhibited by PL cells, a phenomenon that may in part contribute to the preservation of PL width (Ogiso et al., 1991). Various experimental and clinical perturbations, including application of mechanical forces (Andreasen and Schwartz, 1986), heat (Line et al., 1974), and bisphosphonates (Wesselink and Beertsen, 1994) have been used to study PL width homeostasis. These interventions rely in part on the depletion of PL cell populations. Currently it is unclear how PL cell populations can restore their cellular domains after wounding or after application of heavy applied forces. Both of these types of injuries may delete PL cells and encourage bone ingrowth. The periodontal window wound model developed by Melcher (1970) and modified by Gould et al. (1980) facilitates studies of PL homeostasis because precise portions of the alveolar bone and PL can be reproducibly deleted. A synchronous cohort of proliferating cells is generated shortly after wounding (Gould et al., 1980) and the newly generated cells exhibit a defined sequence of differentiation steps that lead to complete regeneration following surgery (Lekic et al., 1996a). Selective deletion of the PL and alveolar bone after wounding causes a transient disruption of the cellular domains required to preserve homeostasis, thereby providing a system to study the regulation of osteogenic cells by adjacent PL cells. Further, the wounding procedure excludes bacteria and epithelial cells, thereby simplifying the model system. As bisphosphonates have been shown to induce ankylosis in mouse molar teeth (Wesselink and Beertsen, 1994), we sought to identify cell populations that are involved in the regulation of PL width in normal and regenerating periodontium. The bisphosphonates of therapeutic importance are synthetic analogues of inorganic pyrophosphate, an endogenous regulator of bone metabolism that can inhibit bone resorption and mineralization in vitro (Fleisch, 1991). All bisphosphonates bind tightly to hydroxyapatite, but unlike pyrophosphate are resistant to metabolism by endogenous phosphatases (Schenk et al., 1986). In animals with experi-

mentally induced periodontitis, treatment with low doses of bisphosphonates reduces loss of bone density (Brunsvold et al., 1992) and preserves the height of the alveolar bone (Quartuccio et al., 1994). In this report we examined the effect of HEBP on the proliferation and differentiation of cells that repopulate alveolar bone and PL using the window wound model. We labelled proliferating cells with 3H-thymidine and used immunohistochemistry to identify in vivo the expression of osteopontin (OPN) and bone sialoprotein (BSP), in situ hybridization to assess mRNA (BSP) and flow cytometry to determine in vitro the presence of OPN in differentiating PL cells. MATERIALS AND METHODS Wound Model and HEBP Treatment

Twenty Wistar male rats weighing 110–130 g on the day of surgery were caged in pairs, fed water and food ad libitum, and were kept in a room with a 12-hour light/dark cycle. Surgery was performed between 10:00 and 12:00 on the third day after the arrival of animals. Periodontal window wounds (,1.2 mm in diameter) were created on the mesial root of the mandibular first molar under general anaesthesia (Halothane, N2O) as described (Gould et al., 1977; Lekic et al., 1996a) (Fig. 1). Following the surgery, groups of animals (N 5 5 per group) were treated daily with HEBP subcutaneously at a dose of 15 mg/kg of body weight beginning immediately after surgery or given vehicle only (control; Fig. 2). Group H1 was treated continuously with HEBP for 1 week and then killed by N2O asphyxiation. Group H1R1 was treated continuously with HEBP for 1 week, after which the treatment was discontinued and animals were allowed 1 week to recover. Rats in the two control groups were killed at 1 week (C1) or 2 weeks (C2) following wounding. An hour prior to sacrifice, all rats were injected intraperitoneally with 2 ml of phosphate buffered saline (PBS; pH 7.4) containing 3Hthymidine (specific activity 5 20 Ci/mmol; NEN, Oakville, ON, Canada) at 1 µCi/g body weight. The mandibles were removed immediately after sacrifice and prepared as described previously for histomorphometric and autoradiographic assessment (Lekic et al., 1996b). Mouse monoclonal anti-rat antibodies (Hybridoma Bank) were used for immunohistochemical localization of BSP and OPN as described previously (Langille and Solursh, 1990, Frank et al., 1993) and their specificity further confirmed by immunoprecipitation of the respective proteins from radiolabeled rat bone cell cultures (Kasugai et al., 1991). Control slides were treated with an irrelevant antibody (anti-human CD4 lymphocyte antigen; Coulter Electronics, Burlington, ON, Canada). The intensity of OPN and BSP staining was classified by visual inspection as intense (3), moderate (2), weak (1), or negative (0) in relation to the expression of these proteins in the surrounding, intact tissue. The localization of proteins was studied in the alveolar bone and in the PL. Graphical depiction of these data was obtained from assessment of the mean staining intensity from at

DIFFERENTIATION OF PERIODONTAL LIGAMENT CELLS

331

considered to be labeled if more than five silver grains overlaid its nucleus (P , 0.001) (McCulloch et al., 1989). Counts of labelled and unlabelled cells in four sites (62,500 µm2 each site) from each section were obtained for each slide to assess proliferation in each site. Counts of total cells were not included in the text due to variations between examined sites as described in a previous study (McCulloch et al., 1989). The areas for analysis were as follows: wounded alveolar bone (site 1); wounded PL adjacent to the tooth (site 2); PL on the unwounded side of the periodontium (site 3); alveolar bone on the unwounded side (site 4). The labelling index for each area was calculated as follows: LI 5 number of labelled cells/number of total cells 3 100. In Situ Hybridization

Fig. 1. A drawing of a longitudinal section through rat molar periodontium to illustrate the location of the wound and the unwounded sides as well as the relative size of the wound. Alveolar bone (AB), cementum (C), pulp (P), dentin (D).

least nine sections from three different animals for each time and site. Radioautography and 3H-thymidine Labelling

Immunostained slides were dipped in full-strength Kodak NTB-2 emulsion; randomly distributed to lighttight, dry boxes; and exposed for 2 weeks at 4°C, as described in previous studies (McCulloch et al., 1989, Lekic et al., 1996b). After exposure, the slides were developed in Kodak D-19 developer and stained through the emulsion with haematoxylin and eosin. The middle section in each 10-section ribbon was examined with a light microscope (Laborlux K, Leitz, Wetzlar, Germany) at a magnification of 3250. All measurements were made with an intraocular grid system (250 µm 3 250 µm final magnification); 625 µm2 each. A cell was

Digoxigenin (DIG)-labelled RNA probes for BSP were prepared using a DIG RNA labelling kit (Boehringer Mannheim Biochemica) according to the manufacturer’s instructions. A 912 bp fragment of rat BSP (Chen et al., 1992), subcloned into the EcoR-I site of pBluescript KS (Stratagene), was obtained from Dr. J. Sodek (University of Toronto, ON). The probes were transcribed from 0.5–1 µg of linearized plasmid in a 20 µl reaction volume using T7 (Sense stand, linearized with EcoRV) or T3 (antisense strand, linearized with SmaI) RNA polymerases. Transcripts were resuspended in 100 µl sterile water containing 20 units of RNAse inhibitor. The optimal probe dilution (1:6.25) was empirically determined in preliminary experiments. In situ hybridization was performed as previously described (Krasnoshtein and Buchwald, 1996) with the exception that DTT was omitted from all post-hybridization steps. Hybridization was carried out overnight at 50°C in humidified chamber (Chen et al., 1994; Arai et al., 1995). Hybridized probes were detected using DIGNucleic Acid Detection Kit (Boerhinger Mannheim) according to the supplier’s protocol, with the exception that sections were incubated for 1 hour with a 1:500 dilution of antibody and overnight with substrate solution containing 2 mM Levamisole and 10% polyvinylalcohol. The latter reagent enhances the color reaction (DeBlock and Debrouwer, 1993). Sections were counterstained with eosin and mounted in 100% glycerol. Localization and the intensity of the in situ hybridization signal on slides were assessed qualitatively using the same procedure as for immunohistochemistry analysis (see above). Morphometric Assessment of Tissues

Three immunostained and radiolabeled sections from each animal were analyzed morphometrically. Image analysis (Leica, Wetzlar, Germany; Bioquant, R&M Biometrics; Nashville, TN) was used to estimate the width of the PL. The width of the PL was measured at the middle of the drilled site and on the opposite, unwounded side. We measured PL width at the root apex separately, as the apical portion of the PL exhibits tissue which is enriched with neural elements and does not contain the well-oriented fibre systems or the same types of cells as the more coronal levels of the PL. Cell counts in the PL for the whole unwounded side were estimated from direct measurements of cell density and

332

P. LEKIC ET AL.

Fig. 2. A schematic diagram of the experimental interventions, timing of wounding and administration of the bisphosphonate (HEBP).

the area of the PL: cells/µm2 3 area of unwounded PL in µm2 5 number of cells in the unwounded PL. Bone width was measured as a perpendicular line to the tooth axis at the level of the junction between the cellular and acellular cementum and from the boundary of the alveolar bone and the PL to the boundary of the connective tissue underlying the oral mucosa. Bone width could not be measured reliably on the wounded side due to the fact that the wound confounded this assessment. In regard to the width measurements of the wound, it was not possible to perform morphometric measurements at 1 week because of poor definition of bone margins at this stage. The width of the regenerating alveolar bone at the drilled site was measured after digitizing the area of the OPN or BSP stained tissue in the bone compartment of the wound. Expression of OPN by PL Cells In Vitro

To examine the effect of HEBP on PL cell expression of this putative osteogenic cell marker (i.e., OPN), PL cells were obtained from the apical half of rat molar roots after tooth extraction as described in Matsuda et al. (1992) with some modifications. At confluence (passage 0), the cells were treated with 0.15% trypsin in 0.5 mM EDTA and subcultured at a 1:3 ratio into 35 mm culture dishes. Pulp cells could have initially contaminated the PL-derived cell cultures but such cells, due to their low proliferative capacity do not predominate in the cultures under the conditions that we have used. Second passage PL cells were treated initially for a period of 7 days with 50 µM HEBP (2 T-75 flasks) or 200 µM HEBP (2 T-75 flasks) in regular media (DMEM). These concentrations and the timing of administration were established in pilot investigations and from earlier experiments (Tenenbaum et al., 1992) which indicated that doses of up to 75 µM were well tolerated by bone cells. However, as high doses of HEBP were used in the in vivo studies, we elected to assess the effects of higher doses in vitro (200 µM). To mimic the HEBP treatment regimens which were used in vivo, some HEBP treated cells (50 µM and 200 µM) were treated

for 1 week and then allowed to recover (DMEM without HEBP) for another week while other cells were treated continuously for 2 weeks. During this period control cells were cultured in DMEM with a change of medium at day 7. At the end of the 2 week incubation period, treated and control PL cells were washed with Ca21/ Mg21-free PBS for 30 second to remove debris and dead cells. Cells were detached with 0.15% trypsin and 0.5 mM EDTA and transferred to a tube. Single cell suspensions were obtained by gentle pipetting, and cells were fixed with an equal volume of 2% paraformaldehyde in a Ca21/Mg21-free PBS at 4°C for 30 min. Fixed cells were pelleted and washed in 0.25% BSA in Ca21/Mg21free PBS. Cells were pelleted, resuspended, and incubated with mouse monoclonal antibody to rat OPN in the same buffer at a dilution of 1:600 for 30 minutes at 4°C followed by 10 minutes at 23°C. Negative controls were incubated for the same period of time in buffer at 4°C with the secondary antibody only. The samples were washed again with a BSA/PBS solution and pelleted. Samples were incubated with FITC-conjugated sheep anti-mouse antibodies diluted 1:100 in the BSA solution and incubated for 30 minutes at 4°C. Stained samples were washed with Ca21/Mg21-free PBS, pelleted and resuspended in 1 ml Ca21/Mg21-free PBS for 10 minutes at room temperature prior to flow cytometry analysis. Flow cytometry was performed on a FACSTAR Plus flow cytometer (Becton Dickinson Immunocytochemistry Systems, Mountainview, CA) equipped with an argon ion laser. FITC was excited by the 488 nm line and emission was collected through a 530/15 interference filter. Forward light scatter was used to include only those particles in the analyses with the expected size of whole cells. Statistical Methods

All slides were coded to permit masked assessment and thereby reduce examiner bias. Raw data were kept separate and the mean for each animal, type of treatment, sacrifice day, and examined site were calculated.

DIFFERENTIATION OF PERIODONTAL LIGAMENT CELLS

The mean from each animal was considered as an independent sample, and these data were then assessed by analysis of variance (ANOVA). Differences between zones were examined by ANOVA and were considered significant at P , 0.05. Data were expressed as mean 6 S.E.M. Post hoc comparisons were made using Duncan’s multiple range test. RESULTS Morphometric Assessment of Tissues

We determined first if HEBP was pharmacologically active by measuring the bone width after treatment. Animals treated with HEBP for 1 week showed twofold greater (P , 0.001) alveolar bone width on the unwounded side compared to controls (Fig. 3A–C). We also determined the longevity of the HEBP effect by treating animals for 1 week with HEBP followed by a 1 week recovery period. After the recovery period the width of the bone on the unwounded side decreased in comparison to bone width at 1 week, but was still considerably thicker than controls (P , 0.001). The intensity of OPN and BSP staining was similar between controls and HEBP-treated animals at 1 week after wounding (Figs. 3F, 4A,B). However, after the recovery period, the immunostaining for OPN in HEBPtreated animals was nearly twofold higher, whereas the control animals showed little change (Fig. 4A). BSP immunostaining increased 9–14 fold in control and HEBP-treated animals, with the increase being significantly higher in the latter group (Fig. 4B; P , 0.001). PL Regeneration

On the unwounded side after 1 week of HEBP treatment, PL width was significantly reduced compared to controls (,40%; P , 0.01; Figs. 3B,C, 5A). Boundaries of the reduced PL were irregular in shape and the PL contained islands of bone-like tissue in the middle of the PL (Fig. 3G). Notably, with serial sectioning, some of the islands were found to be wholly contained within the ligament and did not contact either the bone or cementum margins while other islands were so large that they were actually contiguous with either or both the alveolar bone and cementum (not shown). After the recovery period the PL width increased nearly to that of controls. On the wounded side in animals not treated with HEBP, normal PL width was nearly restored to control levels by 2 weeks after surgery (Fig. 5B). In animals treated with HEBP for 1 week, PL width was reduced significantly (,40% of controls; P , 0.001) and during the recovery period (1 week of HEBP 1 1 week of recovery) the PL width on the wounded side increased and reached control values by the end of the experimental period (Fig. 5B). Thus, independent of the selective deletion of PL cells by wounding, PL width reduction after 1 week of HEBP treatment or the recovery of PL width after 2 weeks was similar in both wounded and non-wounded sides. In contrast to the large difference of PL width between the wounded and unwounded sides, PL width at the apical region of the tooth was similar for all treatment groups (C1 5 189 6 24 µm; C2 5 194 6 36 µm; H1 5 192 6 21 µm; H1R1 5 191 6 29 µm; P . 0.2).

333

HEBP Modulation of PL Cell Differentiation

In control animals, immunostaining for BSP was absent in the PL during the 2 week experimental period (Table 1; Fig. 3B). Staining for OPN in the PL was sparse, detected only at higher magnification (31,000), and was restricted to proliferating cells (Fig. 6C). In animals treated with HEBP, there was strong staining for OPN and BSP throughout the entire experimental period (Table 1). Expression of OPN and BSP in soft tissues was localized specifically within the PL; notably, adjacent gingival connective tissue and pulp were not stained (Fig. 6A,B; BSP). Immunostaining for BSP in the PL was strongly positive after 1 week of HEBP treatment and 1 week of recovery. In situ hybridization showed no BSP mRNA expression in sections prepared with sense probes (data not shown) or in the PL cells of controls (Fig. 6D), but there was a strong signal in animals treated with HEBP (Fig. 6E). The strongest signal for BSP mRNA, as well as for BSP protein (by immunohistochemistry), was in the PL cells of animals treated with HEBP for 1 week followed by 1 week of recovery (Fig. 6E). At the end of the 2-week treatment period, flow cytometry analysis demonstrated that over 50% of the PL cell population was OPN positive (Fig. 7A) and that there was no change in the percentage of positive cells after any of the HEBP treatments. After 2 weeks, the mean OPN fluorescence was the same in control and HEBP-treated cultures regardless of the concentration of HEBP (50 µM or 200 µM HEBP). In contrast, when cultures were treated for 1 week followed by a 1 week recovery, 50 µM and 200 µM HEBP increased the mean OPN expression over 15% and nearly twofold, respectively (Fig. 7B; P , 0.05). Cell Proliferation

Cell counts in unwounded PL of HEBP-treated animals (1 week) were fourfold less than controls (Fig. 8B; P , 0.001). After the recovery period the cell count increased to 84% of control values (Fig. 8B). In parallel with the cell count data, we found that animals treated with HEBP for 1 week showed fivefold lower labelling indices at the unwounded side of the PL compared to controls (Fig. 8E). During the recovery period the labelling indices at the unwounded side of the PL returned to control values. One week wounded controls showed approximately threefold higher proliferation of cells in the extirpated bone area compared to the unwounded side, where proliferation was reduced 30% by 2 weeks (Fig. 8C). HEBP treatment inhibited wound-induced cell proliferation by 10-fold at 1 week, but after the 1-week recovery period the labelling index returned nearly to control levels. We found no difference in the proliferation of cells in the repopulating PL after wounding compared to the unwounded side in animals without HEBP (Fig. 8D,E). However, HEBP inhibited the wound-induced increase in labelling index by threefold at 1 week (Fig. 8D,E). Regardless of the type of treatment, labelling indices in the bone on the unwounded side were close to 0 at both sampling times (Fig. 8, site 4).

Fig. 3.

Page 334

DIFFERENTIATION OF PERIODONTAL LIGAMENT CELLS

335

DISCUSSION

We have studied the effect of a bisphosphonate (HEBP) on the wound healing response of rat periodontium. In addition to the expected increase of alveolar bone matrix (Johnston et al., 1983), perturbation of proliferative responses (Tenenbaum et al., 1992; Goziotis et al., 1995) and reduction of PL width (Wesselink and Beertsen, 1994), we demonstrated that HEBP induced PL cells to express OPN and BSP, welldescribed markers of osteogenic cells. The major focus of research on bisphosphonates relates to their ability to increase bone mass and for treatment of osteoporosis (Valkema et al., 1989), Paget’s disease and hypercalcemia of malignancy (Johnston et al., 1983). HEBP and other bisphosphonates inhibit osteoclast function (Carano et al., 1990; Piper et al., 1994), and this is thought to be the principal mechanism for bisphosphonate-induced increases in bone mass. The central finding of this study is that HEBP can also modulate the differentiation repertoire of soft connective tissue cells. Stimulation of the Osteogenic Phenotype

Consistent with previous work (Lekic et al., 1996b), the PL of controls showed no detectable BSP expression and only very limited OPN expression that was restricted to proliferating cells. In contrast, in animals treated with HEBP, PL cells expressed both OPN and BSP with the highest expression observed during the recovery phase. The present findings demonstrate the recruitment of a significant proportion of the PL cell population to express osteogenic cell characteristics during the recovery phase (i.e., following cessation of treatment with HEBP). As McCulloch et al., (1987) have indicated, cells from the endosteal spaces migrate into the PL and contribute to the PL cell population. Treatment with HEBP could then affect the differentiation of progenitor cells from the endosteal spaces as well as those from the paravascular location in the PL

Fig. 3. A: Histogram of the bone width on the unwounded side of the periodontium (mean 6 S.E.M.). B: Histological section of unwounded (control, Group C2) periodontium. Section was immunostained for BSP (arrowheads). Note the normal width of the periodontal ligament (PL) and the strong staining for BSP in cellular cementum (C) and alveolar bone (AB). Dentine (D) and pulp (P) as well as the PL are not stained for BSP. C: Histological section of unwounded periodontium stained for BSP (arrowheads) after 1 week of HEBP (Group H1). Note that the PL is very narrow and not labeled for BSP. In contrast to the existent alveolar bone (AB), the additional bone matrix exhibits very weak expression for BSP (arrows). Eosin staining was intensive in animals treated 1 week with HEBP. D: Histological section of the wound site after 1 week of HEBP and 1 week of recovery (Group H1R1) immunostained for BSP (arrowheads). Note that the PL is strongly stained for BSP and that there is also staining in the regenerating alveolar bone (RAB) that nearly fills the wound site. E: Histological section of the wound site after 2 weeks without HEBP treatment (Group C2). Section was immunostained for BSP. The cementum (C), original alveolar bone (AB) and regenerating alveolar bone (RAB) are stained for BSP (arrowheads) while the PL is unstained. Note that the width of the PL in this figure is somewhat wider than the width of the PL in figure D in which the animal was treated with HEBP. F: A low power photomicrograph of a wounded site from a 1 week HEBPtreated animal (Group H1). Section was immunostained for BSP. Note that the additional bone matrix (ABM; arrows) is not stained for BSP while the adjacent bone is stained for BSP (arrowheads). G: Histological section of an animal treated with HEBP for 1 week (Group H1). The unwounded PL contains isolated islands of bone matrix (BM) with low levels of BSP and of cellular and morphologically indistinguishable from adjacent cementum (C) and alveolar bone (AB).

Fig. 4. Histograms of osteopontin (OPN) and bone sialoprotein (BSP) staining in regenerating alveolar bone after wounding. Note that OPN staining is enhanced in the recovery period after HEBP compared to controls and that BSP staining is increased 8–10-fold between 1 and 2 weeks after wounding. Data are mean 6 S.E.M.

(McCulloch and Melcher, 1983). Thus HEBP may recruit PL cells to osteoblastic or cementoblastic cell lineages. As shown by immunostaining, the distribution of OPN and BSP followed the exact anatomical borders of the PL and there was no staining found in contiguous gingival and pulpal connective tissues, indicating profound phenotypic differences of cells in PL, gingival and pulpal domains. There are significant variations in phenotype of cells derived from periodontal ligament or gingiva (Mariotti and Cochran, 1990), and gingival fibroblasts cannot participate in regeneration of periodontal ligament (Boyko et al., 1981). The present findings are consistent with the lineage restriction of gingival cells and their inability to switch to a PL phenotype. To determine whether cultured PL cells were capable of expressing OPN and BSP, we used flow cytometry to detect OPN in HEBP-treated PL cells as well as in situ hybridization to demonstrate the expression of BSP mRNA. Expression of OPN in vitro increased during the recovery period, and at 200 µM HEBP there was almost a twofold increase of intracellular OPN. Further, analysis of tissue sections by in situ hybridization

336

P. LEKIC ET AL.

it is not likely that such islands would be wholly enclosed by the PL, and we argue that these findings indicate changes in phenotypic expression which are consistent with the data on OPN and BSP expression by PL cells. Notably, BSP expression was also inhibited by HEBP, a finding that is consistent with putative inhibition of mineralization. Finding of the BSP expression by PL cells is novel, since it was previously shown that only OPN is detectable transiently in the early stages of PL wounding (Lekic et al., 1996a). Collectively, these data indicate that HEBP induces profound phenotypic changes in the PL cells. Inducible cells that can enter the osteogenic lineage are apparently distributed throughout the width of the PL, but in physiological situations these cells are blocked from producing bone and cementum by separate, osteogenic-inhibitory PL cell subpopulations (Ogiso et al., 1991). Presumably, it is the balance of osteogenic inhibitory and osteogenic cell populations that maintains the PL width. HEBP-Induced PL Cell Recruitment

Fig. 5. Histograms of periodontal ligament width. Data (% mean 6 S.E.M.) were estimated from linear measurements of PL width made at the unwounded (A) and wounded (B) sides of the PL and divided by measurements of PL width made at unwounded sites without wounding or HEBP treatment.

showed strong expression for BSP mRNA by large numbers of PL cells after the recovery phase. Previous studies have shown that OPN is expressed by osteogenic cells and also by non-osteogenic cells (Denhardt and Guo, 1993), whereas BSP is expressed almost exclusively by differentiated mineralized tissue-forming cells (Bianco et al., 1991; Chen et al., 1991, 1992). Notably, BSP immunostaining was very weak in alveolar bone that was formed in animals receiving HEBP during the first week following wounding, suggesting that this was an HEBP-mediated event. We found that islands of bone were formed only within the PL of HEBP-treated animals. Since serial sections indicated that some of the islands were wholly contained within the PL, this supports the contention that PL cells were induced to express the osteogenic phenotype. It might be argued that HEBP-induced bone-islands could have been formed as a result of this drug’s inhibitory effects on osteoclastic activity (Carano et al., 1990; Flanagan and Chambers, 1991). However,

The changes in OPN and BSP staining within the PL as well as changes in its width following HEBP treatment suggested that the drug not only altered the differentiation repertoire of PL cells but may have also created selection pressure for specific osteogenic cell lineages already present in the whole PL cell population. We performed detailed analyses of cell counts and proliferation in the periodontium to assess this possibility. Similar to previous studies (Gould et al., 1980; Lekic et al., 1996b), there were rapid and large increases in the percentage of 3H-thymidine labelled cells in wounded but untreated controls at the wound site. These studies have also shown that there were no changes in cell proliferation on the unwounded side of treated animals and thus we conclude that wounding had little or no effect on cell behaviour at the contralateral side. Treatment with HEBP for 1 week strongly inhibited PL cell proliferation and reduced cell counts, consistent with the results of HEBP on chick periosteal cells in vitro (Goziotis et al., 1995) and in PL cell cultures. After the recovery period, experimental animals exhibited a

Fig. 6. A: BSP immunostained section of the unwounded side at the junction between periodontal ligament (PL) and gingival connective tissue (G). Animal was treated with HEBP for 1 week and then allowed to recover for 1 week (Group H1R1). Note that BSP staining (arrowheads) is restricted to the PL and alveolar bone (AB) and follows the anatomical borders of the PL. Dentine (D), periodontal ligament (PL), cementum (C). B: BSP immunostained section of periodontium at the wounded side close to the apex of tooth after 1 week HEBP and 1 week recovery (Group H1R1). Alveolar bone (AB), periodontal ligament (PL) and the junction of cementum (C) and dentine (D) are strongly stained (arrowheads) whereas dentine, pulp (P) and apical tissues (A) are unstained. C: Combined radioautography and immunostaining for OPN at a wounded site. Animal was wounded (Group C2) 2 weeks before injection with 3H-thymidine and then killed one hour later. Note the spatial association of OPN staining (arrowheads) and 3H-thymidine labelled cell (arrows). D: In situ hybridization of BSP antisense-RNA probe to mRNA in cells on surface of cementum but there is no signal in PL cells (unwounded side). Animal was not treated with HEBP (Group C2). E: In situ hybridization of BSP antisense-RNA probe to mRNA in cells of periodontal ligament (PL) at the unwounded side after 1 week of HEBP and 1 week of recovery (Group H1R1). Note that cells in the PL exhibit strong signals for BSP expression.

Fig. 6.

Page 337

338

P. LEKIC ET AL.

marked increase in cell counts and labelling indices, indicating homeostatic regulation of PL cell populations. The largest increases in cell proliferation after the recovery period were in the body of the PL and in the region adjacent to the alveolar bone, the same sites that exhibited strong staining for BSP. Our in vitro studies showed that during the first and second days of HEBP treatment (200 µM), a large portion of PL cells died, but that the remaining cells proliferated avidly thereafter and expressed higher levels of OPN. CollecTABLE 1. Summary description of bone marker expression in PL at the unwounded side of HEBP treated animals and controls Duration and type of treatment 1 week Bone markers Controls OPN BSP

0 0

HEBP 1.2 6 0.2 1.4 6 0.2

2 weeks Controls 1 wk HEBP 1 1 wk recovery 0 0

2.0 6 0.0 3.0 6 0.0

Data presents mean (6s.e.m.) OPN and BSP expression obtained by visual inspection of immunostained sections following the different type of treatment and post-wounding time.

Fig. 7. A,B: Histograms of the % of OPN1ve cells and mean intracellular OPN fluorescence intensity in rat periodontal ligament cell cultures treated as indicated in legend. OPN1ve cells were assessed by flow cytometric analyses and calculation of background fluorescence from cells stained with second antibody only. The intracel-

tively these data indicated that the surviving cells were resistant to the cytocidal effects of HEBP and that HEBP treatment presumably selects for PL cells that enter the osteogenic lineage. In particular, trauma and application of physical force induces PL cells to proliferate, migrate, and differentiate into osteoblasts (Roberts and Chase, 1981); and it has been suggested that about 50–60% of PL cells may have osteogenic potential (Roberts and Chase, 1981; Yee et al., 1976; Yee, 1979). Our study is in general agreement with these previous findings and demonstrates that a bisphosphonate could have a potentially therapeutic use in recruiting PL cells to a bone-forming phenotype, thereby promoting early formation of alveolar bone and cementum in wound healing. ACKNOWLEDGMENTS

We thank Wilson Lee for assistance with flow cytometry, Harry Moe for the work in cell culturing, Ron Zohar for help with immunostaining, Caroline Chu for preparation of the manuscript, Peter Fritz for the HEBP treatment, Ian Speers for general assistance, and Jaro Sodek for advice. C.M. and H.T. are supported by an MRC of Canada Group Grant. S.C. is supported by an MRC of Canada Scholarship.

lular staining intensity of OPN is expressed as mean fluorescence channel number. Treatment of cells with HEBP at 200 µM for 1 week and 1 week recovery induced a large and significant increase of OPN expression compared to controls (P , 0.05).

DIFFERENTIATION OF PERIODONTAL LIGAMENT CELLS

Fig. 8. A: Diagram of cross-sectional view through periodontium to illustrate the location of sample sites for proliferation analyses. B: Cell counts for whole periodontal ligament (mean 6 S.E.M.). Note that 1 week of HEBP treatment reduces cell counts by fourfold and that after 1 week of recovery, cell counts are nearly that of controls. C: Labelling indices of cells in repopulating alveolar bone after extirpation (site 1). Note that HEBP reduces proliferation by 10-fold. Mean 6 S.E.M. D: Labelling indices of cells that repopulate the PL space (site 2).

339

Proliferation is much lower than in the repopulating alveolar bone site. HEBP reduces labelling indices at 1 week by threefold but cells recover after 1 week without HEBP. Mean 6 S.E.M. E: Labelling indices at unwounded PL (site 3). HEBP strongly inhibits proliferation at 1 week but cells recover after 1 week without HEBP. Mean 6 S.E.M. F: Labelling indices in unwounded alveolar bone (site 4). For all treatments, labelling indices are close to 0. Mean 6 S.E.M.

340

P. LEKIC ET AL.

LITERATURE CITED Andreasen, J.O. and O. Schwartz 1986 The effect of saline storage before replantation upon dry damage of the periodontal ligament. Endocr. Dent. Trauma., 2:67–70. Arai, N., K. Ohya, S. Kasugai, H. Shimokawa, S. Ohida, H. Ogura, and T. Amagasa 1995 Expression of bone sialoprotein mRNA during bone formation and resorption induced by colchicine in rat tibial bone marrow cavity. J. Bone Miner Res., 10:1209–1217. Bianco, P., L.W. Fisher, M.F. Young, J.D. Termine, and P.G. Robey 1991 Expression of bone sialoprotein (BSP) in developing human tissues. Calcif. Tissue Int., 49:421–426. Boyko, G., A.H. Melcher, and D.M. Brunette 1981 Formation of new periodontal ligament by periodontal ligament cells implanted in vivo after culture in vitro. J. Periodont. Res., 16:73–88. Brunsvold, M.A., E.S. Chaves, K.S. Kornman, T.B. Aufdemorte, and R. Wood 1992 Effects of a bisphosphonate on experimental periodontitis in monkeys. J. Periodontol., 63:825–830. Carano, A., S.L. Teitelbaum, J.D. Konsek, P.H. Schlesinger, and H.C. Blair 1990 Bisphosphonates directly inhibit the bone resorption activity of isolated avian osteoclasts in vitro. J. Clin. Invest., 85:456–461. Chen, J., H.S. Shapiro, J.L. Wrana, S. Reimers, J.N.M. Heersche, and J. Sodek 1991 Localization of bone sialoprotein (BSP) expression to the site of mineralized tissue formation in fetal rat tissues by in situ hybridization. Matrix, 11:133–143. Chen, J., H.S. Shapiro, and J. Sodek 1992 Developmental expression of bone sialoprotein mRNA in rat mineralized connective tissues. J. Bone Miner. Res., 8:987–997. Chen, J.K., M.D. McKee, A. Nanci, and J. Sodek 1994 Bone sialoprotein mRNA expression and ultrastructural localization in fetal porcine calvarial bone: comparison with osteopontin. Histochem. J., 26:67–78. DeBlock, M. and D. Debrouwer 1993 RNA-RNA in situ hybridization using digoxigenin-labelled probes: the use of high molecular weight polyvinyl alcohol in the alkaline phosphatase indoxylnitroblue tetrazolium reaction. Anal. Biochem., 215:86–89. Flanagan, A.M. and T.J. Chambers 1991 Inhibition of bone resorption by bisphosphonates: interactions between bisphosphonates, osteoclasts and bone. Calcif. Tissue Int., 49:407–415. Fleisch, H. 1991 Bisphosphonates: Pharmacology and use in the treatment of tumour-induced hypercalcaemic and metastatic bone disease. Drugs, 42:919–944. Frank, J.D., R. Balena, P. Masarachia, J.G. Seedor, and M.E. Cartwright 1993 The effects of three different demineralization agents on osteopontin localization in adult rat bone using immunohistochemistry. Histochemistry, 99:295–301. Gould, T.R.L., A.H. Melcher, and D.M. Brunette 1977 Location of progenitor cells in periodontal ligament of mouse molar stimulated by wounding. Anat. Rec., 188:133–141. Gould, T.R.L., A.H. Melcher, and D.M. Brunette 1980 Migration and division of progenitor cell populations in the periodontal ligament after wounding. J. Periodont. Res., 15:20–42. Goziotis, A., B. Sukhu, M. Torontali, M. Dowhaniuk, and H.C. Tenenbaum 1995 Effects of bisphosphonates APD and HEBP on bone metabolism in vitro. Bone, 16:3175–3275. Johnston, C.C., R.D. Altman, R.E. Canfield, G.A.M. Finerman, J.D. Taulbee, and M.C. Ebert 1983 Review of fracture experience during treatment of Paget’s disease with etidronate disodium (EHDP). Clin. Orthop., 172:186–194. Kasugai, S., R. Todescan, T. Nagata, K.L. Yao, W.T. Butler, and J. Sodek 1991 Expression of bone matrix proteins associated with mineralized tissue formation by adult rat bone marrow cells in vitro: inductive effects of dexamethasone on the osteoblastic phenotype. J. Cell. Physiol., 147:111–120. Krasnoshtein, F. and M. Buchwald 1996 Developmental expression of the FAC gene correlates with congenital defects in Fanconi anemia patients. Human. Mol. Genetics, 5: 85–93.

Langille, R.M. and M. Solursh 1990 Formation of chondrous and osseous tissues in micromass cultures of rat frontonasal and mandibular ectomesenchyme. Differentiation, 44:197–206. Lekic, P., J. Sodek, and C.A.G. McCulloch 1996a Osteopontin and bone sialoprotein expression in regenerating rat periodontal ligament and alveolar bone. Anat. Rec., 244:52–58. Lekic, P., Sodek, J., and C.A.G. McCulloch 1996b Relationship of cellular proliferation to expression of osteopontin and bone sialoprotein in regenerating rat periodontium. Cell Tissue Res., in press. Line, S.E., A.M. Poslon, and H.A. Zander 1974 Relationship between periodontal injury, selective cell repopulation and ankylosis. J. Periodontol., 45:725–30. Mariotti, A. and D.L. Cochran 1990 Characterization of fibroblasts derived from human periodontal ligament and gingiva. J. Periodontol., 60:103–11. Matsuda, N., N.M. Kumar, P.R. Ramakrishnan, W-L Lin, R.J Genco, and M.I. Cho 1993 or 92 Evidence for upregulation of epidermal growth-factor receptors on rat periodontal ligament fibroblast cells associated with stabilization of phenotype in vitro. Arch. Oral Biol., 38:559–69. McCulloch, C.A.G. and A.H. Melcher 1983 Cell density and cell generation in the periodontal ligament of mice. Am. J. Anat., 167:43–58. McCulloch, C.A.G., E. Nemeth, B. Lowenberg, and A.H. Melcher, 1987 Paravascular cells in endosteal spaces of alveolar bone contribute to periodontal ligament cell populations. Anat. Rec., 219:2233– 2242. McCulloch, C.A.G., U. Barghava, and A.H. Melcher 1989 Cell death and the regulation of populations of cells in the periodontal ligament. Cell Tissue Res., 255:129–138. Melcher, A.H. 1970 Repair of wounds in the periodontium of the rat. Influence of periodontal ligament on osteogenesis. Arch. Oral Biol., 15:1183–1204. Melcher, A.H. 1976 On the repair potential of periodontal tissues. J. Periodontol., 47:256–260. Ogiso, B., F.I. Hughes, A.H. Melcher, and C.A.G. McCulloch 1991 Fibroblasts inhibit mineralized bone nodule formation by rat bone marrow stromal cells in vitro. J. Cell. Physiol., 146:442–450. Piper, K., A. Boyde, and S.J Jones 1994 The effect of 3-amino-1hydroxypropylidene-1, 1-bisphosphonate (APD) on the resorptive function of osteoblasts of known nuclear number. Calcif. Tissue Int., 54:56–61. Quartuccio, W.M., J.G. Seedor, T.B. Aufdemorte, M. Brunsvold, E. Chaves, K.S. Kornman, and G.A. Rodan 1994 Histomorphometrical analysis of the effects of the bisphosphonate alendronate on bone loss caused by experimental periodontitis in monkeys. J. Periodont. Res., 29:35–40. Roberts, W.E. and D.C. Chase 1981 Kinetics of cell proliferation and migration associated with orthodontically-induced osteogenesis. J. Dent. Res., 60:174–181. Schenk, R., P. Eggli, H. Fleish, and S. Rosini 1986 Quantitative morphometric evaluation of the inhibitory activity of new aminobisphosphonates on bone resorption in the rat. Calcif. Tissue Int., 38:342–349. Tenenbaum, H.C., M. Torontali, and B. Sukhu 1992 Effects of bisphosphonates and inorganic pyrophosphate on osteogenesis, in vitro. Bone, 13:249–255. Wesselink, P.R. and W. Beertsen 1994 Ankylosis of the mouse molar after systemic administration of 1-hydroxyethylindene-1, 1-bisphosphonate (HEBP). J. Clin. Periodont., 21:465–471. Yee, J.A., D.B Kimmel, and W.S.S. Jee 1976 Periodontal ligament cell kinetics following orthodontic tooth movement. Cell Tissue Kinet., 9:293–302. Yee, J.A. 1979 Response of periodontal ligament cells to orthodontic force: ultrastructure identification of proliferating fibroblasts. Anat. Rec., 194:603–614.