Mechanical forces exerted on teeth are transmitted to the tissues surrounding their roots, and remodelling activity is produced in response. The mechanisms regulating orthodontic movements are not clearly understood, but several studies have pointed out the important role of prostaglandin biosynthesis in force-induced tooth movement. Increased amounts of PGE have been demonstrated immunohistochemically in cells in close vicinity to the tooth after orthodontic movements in the cat . In line with this observation, local injections of prostanoids during orthodontic movement concentration-dependently increased the number of osteoclasts in the rat and significantly accelerated the rate of the movement in man . Furthermore, indomethacin, an inhibitor of cyclo-oxygenase, reportedly decreased experimental tooth movement in the cat . However, the mechanisms that lead to increased prostanoid biosynthesis in tissues surrounding the teeth, including the periodontal ligament, are not understood. Inflammatory reactions, such as increased numbers of phagocytic cells, increased activity of macrophages, local vascular injuries and tissue degradation, have been demonstrated in periodontal tissues following orthodontic movement .We have recently reported that the inflammatory mediators bradykinin and thrombin stimulate the release of arachidonic acid and the formation of prostanoids in cells of the human periodontal ligament in vitro and consequently may contribute to prostanoid formation associated with orthodontic movements .

Inflammatory cytokines such as IL-1, IL-1?, and TNF have been immunolocalized in the paradental tissues of cat canines subjected to orthodontic forces . Accordingly, recent studies have shown that PGE, IL-1? and TNF are increased in human gingival crevicular fluid during orthodontic tooth movement . Furthermore, cultured human periodontal-ligament cells respond to IL-1, IL-1? and TNF with increased production of PGE .

Several cytokines and inflammatory mediators are produced concomitantly in inflammatory reactions, including that in the paradental tissues during orthodontic movements. Synergistic, interactive effects on prostaglandin biosynthesis between bradykinin and different cytokines have been demonstrated in human gingival fibroblasts (Lerner and Modeér, 1991) and human dental-pulp fibroblasts. However, periodontal tissues harbour several cell types, which implies that the cells of the ligament are not only fibroblasts. Progenitor cells of osteoblasts, cementoblasts as well as fibroblasts exist in the ligament . Furthermore, cells of the human periodontal ligament, but not gingival fibroblasts, were capable of producing mineral-like nodules in vitro, indicating differences in behaviour between gingival fibroblasts and ligament cells.

The findings of synergistic effects of inflammatory mediators on prostaglandin biosynthesis in human gingival or pulpal fibroblasts cannot be taken as evidence for the same reactions in cells of the periodontal ligament. We have therefore now investigated possible interactions between bradykinin, thrombin and IL-1, IL-1? and TNF-, TNF-? on prostanoid formation and the release of arachidonic acid from cultured human periodontal-ligament cells.

-MEM was purchased from GIBCO-Life Technologies Ltd, Paisley, Scotland; FCS from Flow Laboratories, Irvine, Scotland; bacterial collagenase (Clostridium type 1) from Worthington Biochemical Freehold, NJ, U.S.A.; bradykinin from Sigma Chemical Co., St. Louis, MO, U.S.A.; recombinant human IL-1 (spec. act. 10⁸ U/mg), recombinant human IL-1? (spec. act. 5×10⁸ U/mg), recombinant human TNF- (spec. act. 2×10⁷ U/mg), recombinant human TNF-? (spec. act. 3×10⁷ U/mg) from Genzyme, Boston MA, U.S.A.; [³H]arachidonic acid and the radioimmunoassay kits for PGE₂ and 6-keto-PGF₁ from Du Pont/New England Nuclear Chemicals, Dreieich, Germany; multiwell plastic culture dishes from Costar, Cambridge, MA, U.S.A.; thrombin (topostatin) from Hoffman-La Roche & Co, Switzerland.

Healthy teeth were extracted from three patients in the course of orthodontic treatment. The crowns of the teeth were carefully cleaned and the teeth washed in Tyrode’s salt solution with antibiotics (100 IU/ml pencillin; 50 ?g/ml streptomycin) added. Cells of the periodontal ligament were isolated from the roots of the teeth by collagenase digestion, as described in detail . The cells released were collected, centrifuged, resuspended in -MEM with 10% FCS, and seeded in culture flasks. The cells were grown at 37°C in a humidified atmosphere of 95% air and 5% CO₂ in -MEM supplemented with 10% FCS, 100 IU/ml pencillin, 50 ?g/ml streptomycin, and 0.7 mM -glutamine.

The medium was changed every third day. Cells from passages 3–9 were used in the experiments. Cells were seeded in multiwell dishes (2 cm²) and incubated overnight, as described .

Before the experiments the cells were rinsed twice with prewarmed (37°C) Tyrode’s solution and preincubated for 30 min in -MEM without FCS. The preincubation medium was then removed and fresh medium with test substances or vehicle was added for the stated times and at the stated concentrations. At the end of the experiments, media were withdrawn, acidified to pH 3.5, frozen and stored at ?20°C for later analysis of PGE₂ and 6-keto-PGF₁. After the experiments the cells were detached with trypsin (1 mg/ml) and counted in a haemocytometer. Cell densities at the end of experiments is stated in the figure legends.

The amounts of PGE₂ and prostacyclin (as assessed by analysis of the stable breakdown product, 6-keto-PGF₁) in the media were analysed with commercially available radioimmunoassay kits, using [¹²⁵I]PGE₂ and [¹²⁵I]6-keto-PGF₁ as tracers.

Subconfluent cells were rinsed once with serum-free medium and then incubated in serum-free medium containing 0.5 ?Ci [³H]arachidonic acid/ml. After 24 hr, the medium was withdrawn and the cells washed five times with serum-free medium. Subsequently, 0.5 ml serum-free, HEPES-buffered (20 mmol/l) medium with or without test substances was added and cells were incubated for 30 min. Thereafter, samples of media were withdrawn and analysed for ³H in a liquid scintillation counter. The activity found in the supernatant represented free [³H]arachidonic acid as well as [³H]-labelled metabolites.

The addition of bradykinin (1 ?mol/l) and IL-1? (10 U/ml) to the human periodontal-ligament cells in vitro resulted in a time-dependent stimulation of PGE₂. The effect of bradykinin was significant after 7.5 min (the first time-point studied), with maximal effect at 15 min . The effect of IL-1? on prostanoid biosynthesis was more delayed than that of bradykinin. A small but significant effect of IL-1? on PGE₂ formation was demonstrated after 30 min and maximal stimulation was seen after 8 hr . Il-1 and IL-1?, at and above 3 and 0.3 U/ml, respectively, concentration-dependently enhanced PGE₂ biosynthesis with optimal effects at 10 U/ml . When bradykinin (1 ?mol/l) was added together with different concentrations of Il-1 or IL-1?, a synergistic, concentration-dependent potentiation (more than an additive effect) of PGE₂ formation was demonstrated in 24-hr cultures . A synergistic and concentration-dependent interaction between bradykinin and Il-1 or -1? on 6-keto-PGF₁ biosynthesis was also demonstrated in 24-hr cultures .

Statistical analysis was done with student’s t-test for unpaired samples. Experiments were repeated two or three times. Figures show representative experiments.Fig. 1. (a) Short-term time-course study of the effect of bradykinin (BK) (1 ?mol/l) and interleukin (IL)-1? (10 U/ml) on prostaglandin (PG)E₂ biosynthesis in human periodontal-ligament cells. Points represent means for four wells. SEM is shown as vertical bars. Cell density 10⁴ cells/cm². The effects of BK and IL-1? were statistically significant after 15 (p<0.05) and 30 min (p<0.005), respectively. (b) Long-term time-course study of the effect of IL-1? (10 U/ml) on PGE₂ biosynthesis in human periodontal-ligament cells. Points represent means for four wells. SEM is shown as vertical bars when larger than the radius of the symbol. Cell density 3.5×10⁴ cells/cm².