Tooth replantation disrupts the neurovascular supply to the dental pulp. The ensuing ischaemia affects most types of pulpal cells and thus usually causes progressive pulp degeneration and concomitant necrosis. In teeth with incomplete roots, however, successful pulp-tissue healing is known to occur after replantation. This process involves various cellular events including revascularization, reinnervation, proliferation and differentiation of pulpal progenitor cells and new hard-tissue formation. Clinical and experimental studies have indicated that infection of the traumatized pulp is a major hindrance to pulp regeneration after replantation Shortly after replantation, the injured pulp becomes more susceptible to infection, mainly due to the impaired neurovascular supply, which results in a loss of the protective potential to the traumatized pulp. Little is known about how the pulps of replanted teeth initially defend against bacterial invasion under such infection-prone conditions.

In the traumatized pulp tissues after replantation, macrophages are believed to act as scavenger cells that phagocytose cellular matrix debris and thus clean the wound site to promote healing. Macrophages are now known to possess several properties other than phagocytosis, such as antigen presentation and the secretion of cytokines. They also produce growth factors for fibroblasts and vascular endothelial cells, by which mechanism they could promote the repair and/or regeneration of injured tissues.

Dendritic cells in the dental pulp express class II molecules and act as dedicated antigen-presenting cells, which can uptake, process and present foreign antigens to CD4+ T lymphocytes during the initial phase of the immune response against external protein antigens. Increased accumulation of dendritic cell-like, class II molecule-expressing cells has been reported in rat molars following experimental caries induction, pulp exposure and cavity preparation, as well as in human teeth bearing dentinal caries. These findings suggest that class II molecule-expressing cells are actively involved in the pulpal defence reactions following the invasion of bacterial elements.

The kinetics of pulpal macrophages and class II molecule-expressing cells after tooth replantation has not yet been investigated. Rat molars have thin dentine that lacks an enamel covering at the tip of cusps and thus may provide a good experimental model for investigating how transdentinal bacterial challenge influences the dental pulp. Hence we have now investigated immunohistochemically the temporal changes in the distribution and number of macrophage-associated antigen-expressing cells of the dental pulp in experimentally replanted rat molars. Our aim was to test the hypothesis that these cells respond to transdentinal pathogenic challenges by protecting the injured pulp against infection, and play some part in facilitating pulp-tissue healing.

Specific pathogen-free, 5-week-old, male Wistar rats (Clea Japan, Osaka, Japan) were used. Under sodium pentobarbital anaesthesia (30 mg/kg, i.p.), the right maxillary first molars were replanted according to the method described by Kvinnsland et al. (1991) and Byers et al. (1992). The experimental teeth were extracted with an excavator inserted from the distopalatal angle while part of the mesial attached gingiva was left intact. The teeth were rotated once anteriorly so that all the roots came out of the socket, and then repositioned immediately. No postoperative splinting was used. Vicciline (20 mg/kg, i.p.) was given as a preoperative antibiotic. The unoperated contralateral first molars served as controls.

After 0 h (immediately after the replantation), 8 h, 1, 3, 7, 14, 28, 56 and 84 days (n=3, 6, 6, 12, 12, 9, 8, 8 and 8, respectively), the animals were anaesthetized as described above and perfused with 2% paraformaldehyde, 0.01 M sodium periodate and 0.075 M lysine in 0.0375 M phosphate buffer. The replanted and control teeth with surrounding jaw bones were then removed, postfixed in the same fixative at 4°C for 24 h, demineralized in 14% EDTA at 4°C, and cut serially in a cryostat.

Immunoperoxidase staining was done on 7-?m thick mesiodistal sections by the use of monoclonal antibodies OX6 (anti-class II molecules; Serotec, Blackthorn, UK), ED1 (a pan-macrophage antibody, reactive also with dendritic cells; Serotec) and ED2 (anti-tissue macrophages; Serotec) as primary antibodies. Following incubation with one of the primary antibodies (4°C, 16 h), the sections were sequentially incubated with biotinylated antimouse IgG (rat adsorbed; Vector, Burlingame, CA) for 30 min, 3% H₂O₂ for 5 min, and avidin–biotin–peroxidase complex (Vector) for 30 min. The sections were then developed with a DAB substrate kit (Vector) and counterstained with methyl green. Details of the immunoperoxidase staining are given in our previous reports.

Double immunoperoxidase staining was according to a previously developed protocol with a few modifications. The first antigen to be visualized was stained as described above, except for the use of an SG substrate kit (dark-blue reaction products; Vector) as a chromogen. The second antigen was then stained with peroxidase-conjugated rat antimouse IgG (Jackson, West Grove, CA) as a secondary antibody and an AEC substrate kit (red reaction products; Dako, Glostrup, Denmark) as a chromogen.

Cell quantification was done on a representative section, chosen from the most central cuts through the coronal pulp, for each primary antibody from each specimen that had been subjected to the immunoperoxidase staining. Immunopositive cells in the coronal pulp with a distinct profile and a nucleus were counted in 10 high-power fields (×600) by using a 10×10 mm ocular grid and a ×40 objective. The fields were determined by scanning mesiodistally the whole coronal pulp of each section in a zigzag fashion without any overlap of the fields. Results were expressed as the mean count of cells per 10 high-power fields. Statistical analysis was made by Student’s t-test.

Some specimens were subjected to immunofluorescence staining and confocal laser scanning microscopy. Free-floating sections, 50 ?m in thickness, were prepared, treated with OX6 (dilution 1:500) or ED2 (dilution 1:250) for 24 h at 4°C, and then incubated with fluorescein isothiocyanate-conjugated, horse antimouse IgG (rat adsorbed; Vector) for 1 h. The stained sections were picked up onto glass slides and examined with a confocal laser scanning microscope (LSM 310; Carl Zeiss, Germany). Typically, a series of optical sections was obtained at 0.5–2-?m intervals throughout the depth of the structure of interest. Three-dimensional reconstruction and rotation were done with software supplied with the microscope.

The specificity of the immunostaining was similar to that reported previously. Double immunoperoxidase staining confirmed our previous finding that virtually all OX6+ and ED2+ cells coexpress immunoreactivity to ED1.

Fig. 4. A double-immunostained picture of the upper right first molar of a rat 7 days after tooth replantation. ED1+ cells (dark blue) and OX6+ cells (red) in the central portion of the distal pulp horn are shown. Arrows, ED1+/OX6? cells. Arrowheads, double-positive cells. Bar=20 ?m.

As reported previously, ED1+ cells and ED2+ cells were distributed throughout the entire pulp while OX6+ cells were concentrated in and around the odontoblastic layer of the coronal pulp. The number of these cells gradually increased with time, reflecting the age-related increase reported elsewhere.

Fig. 1. Changes in the density of immunocompetent cells in the coronal pulp of the replanted teeth: ? experimental teeth; ? control teeth. Each value represents the mean and SD of 3–8 samples. At 14 days, values obtained from reparative dentine-forming pulps and bony pulps were separated and indicated as RD and bone, respectively. At 28 days and later, cell quantification was not done because of extensive calcified-tissue formation. **p<0.01 versus contralateral normal teeth. ^#p<0.05 and ^#p<0.01 versus the immediately previous time-point.