Long-term Hyperglycemia Naturally Induces Dental Caries but Not Periodontal Disease in Type 1 and Type 2 Diabetic Rodents

Periodontal disease (PD) in patients with diabetes is described as the sixth most common complication of diabetes (1). Several recent studies have reported that the severity of PD in patients with type 1 and type 2 diabetes is higher than that in individuals without diabetes (2). Diabetic rodent models subjected to experimental manipulations such as ligature placement have been used as diabetic PD models (3–8). Furthermore, several studies have reported that only long-term hyperglycemia can lead to naturally occurring PD without experimental manipulation in diabetic rats (9,10).

Our previous studies revealed that both PD and dental caries, which have completely different etiologies, apparently occur in these diabetic models (11–13), and the onset and progressive phases of dental caries may be strongly involved in the development of periodontal inflammation. However, we were not able to exclude the possibility that hyperglycemia might induce PD-derived inflammation in diabetic animals. If hyperglycemia induces PD in diabetic animals, then PD-derived inflammation would independently emerge in periodontal tissues by preventing carious inflammation. On the basis of this hypothesis, the current study was designed to assess the utility of alloxan-induced diabetic rats and diabetic db/db mice as potential PD models by protecting dental caries with fluoride treatment, which is considered effective against dental caries not only in humans but also in experimental animals (14,15).

Six-week-old female F344 rats were obtained from Japan SLC, Inc. (Hamamatsu, Japan). Eight-week-old male db/db and db/+ mice were purchased from Charles River Laboratories (Yokohama, Japan). The animals were housed and handled under the same conditions as reported in previous studies (12,13,16,17). All animal experiments were approved by the Committee for Animal Experiments of Setsunan University.

Thirty rats (7 weeks of age) were intravenously administered a single dose (35 mg/kg body weight) of alloxan according to the previous studies (13,17) and were divided into three groups. Each group of 10 diabetic rats was given either tap water alone (the DM F0 group) or tap water containing 10 and 50 ppm sodium fluoride (the DM F10 and DM F50 groups, respectively). Each group of 10 nondiabetic rats (without alloxan dosing) was also given either tap water alone (the non-DM F0 group) or tap water containing 50 ppm sodium fluoride (the non-DM F50 group). The surviving 43 rats were autopsied at 20 weeks of age for morphological examination.

From 10 weeks of age onward, each group of 10 db/db mice was given either tap water alone (db/db F0 group) or tap water containing 25 or 50 ppm sodium fluoride (the db/db F25 and db/db F50 groups, respectively). Simultaneously, each group of 10 db/+ mice was given either tap water alone (db/+ F0 group) or tap water containing 25, 50, or 100 ppm sodium fluoride (the db/+ F25, db/+ F50, and db/+ F100 groups, respectively). The surviving 56 mice were autopsied at 40 weeks of age for morphological examination.

Glucose levels in fresh urine and tail vein blood samples were measured according to the previous studies (12,13,17).

Macroscopic examination of dental caries in mice was performed using a stereoscope under the same conditions as described in a previous study (12).

Soft X-ray examination of dental caries and alveolar bone resorption (ABR) in rats was performed under the same conditions as described in previous studies (13,17).

Histopathological examination was performed on the mandible and maxilla of each rat and mouse. Dental caries including pulpitis (P), apical periodontitis (AP), periodontal inflammation including gingivitis (GV), marginal periodontitis (MP), and ABR were evaluated and graded histopathologically according to the previous study (17).

The χ² test and Wilcoxon rank sum test were used according to the previous studies (12,13,17). A P value <0.05 was considered statistically significant.

In all diabetic alloxan-treated rats and db/db mice, severe hyperglycemia and glucosuria continued during the experimental period. In all nondiabetic alloxan-nontreated rats and db/+ mice, the blood and urine glucose levels were normal.

Result of soft X-ray examination revealed that the development of dental caries was suppressed by fluoride treatment in a dose-dependent manner in diabetic rats, and the lesions were scarcely observed in the DM F50 group. In comparison, almost half of the molars were affected with dental caries in the DM F0 group. Additionally, the mean caries scores and caries incidence in the DM F10 and DM F50 groups were significantly lower (P < 0.01) than those in the DM F0 group. No carious lesions were observed in any molars in the nondiabetic rats (Fig. 1A and Supplementary Table 1).

High mean scores and high incidence of ABR were observed in the DM F0 group using soft X-ray examination; however, bone resorption was markedly suppressed by fluoride treatment, which decreased dental caries. Lesions were not detected in any maxillary molars in the DM F10 and DM F50 groups or in the mandibular molars in the DM F50 group. In the non-DM F0 and non-DM F50 groups, ABR was not observed in any molars (Fig. 1B and Supplementary Table 2).

In addition, ABR was characterized on the basis of its presence in the apical area adjacent to carious molars (Fig. 1C). There were no radiolucent changes in the alveolar bone around the noncarious molars in the fluoride-treated diabetic rats (Fig. 1D).

Histopathologically, fluoride treatment significantly (P < 0.01) suppressed the incidence of dental caries in the diabetic rats. Together with the prevention of dental caries, the incidence of AP was also markedly decreased (P < 0.01) in the DM F10 and DM F50 groups compared with the DM F0 group. Furthermore, MP was not observed in any molars of the DM F10 and DM F50 groups. The incidence of GV in these two groups was significantly decreased (P < 0.01) compared with that in the DM F0 group and was comparable to the incidence in fluoride-treated and -untreated nondiabetic rats (Fig. 2A).

In the fluoride-untreated diabetic rats, MP was always accompanied by moderate (++) caries with P and AP as well as moderate (++) GV. The inflammatory cells in the periodontal tissue (i.e., MP) were exclusively adjacent to suppurative AP (Fig. 2B). Furthermore, neither AP nor MP was detected in any region around the noncarious molars regardless of the presence or absence of diabetes (Fig. 2C and Supplementary Table 3 and 4).

Fluoride treatment also macroscopically suppressed dental caries in the diabetic db/db mice; however, this effect was milder than that in the diabetic rats. Caries development tended to be largely prevented by fluoride treatment, and the mean score and incidence of lesions in the db/db F50 group were significantly lower than those in the db/db F0 group. No dental caries were observed in the db/+ F0 group, although some molars (<10%) were affected with dental caries in the db/+ F25, db/+ F50, and db/+ F100 groups (Fig. 3A and Supplementary Table 5).

Histopathologically, fluoride treatment mildly suppressed the development of dental caries and reduced the incidence of lesions in the db/db F0 group to almost half of that in the db/db F25 and db/db F50 groups. In mice, the incidence of AP was almost the same as that for MP, and the incidence of these two lesions in the db/db F25 and db/db F50 groups was approximately half of that in the db/db F0 group, as was the case with the incidence of caries. GV was equally observed in almost 40% of each group of nondiabetic db/+ mice, whereas the incidence of GV was almost doubled in the db/db F0 group. The incidence of GV in the db/db F25 and db/db F50 groups was slightly suppressed as the rate of dental caries decreased; however, the incidence of GV was not reduced as much as it was in nondiabetic mice with few dental caries (Fig. 3B).

In fluoride-untreated diabetic db/db mice, ABR and MP were inevitably accompanied by moderate caries (++) with P, AP, and moderate GV (++) (Fig. 3C–E). In contrast, ABR and MP were not detected in any region around the noncarious molars regardless of the presence or absence of diabetes (Fig. 3F–H). In addition, ABR and MP remained in the fluoride-treated diabetic db/db mice and were always found together with AP (Fig. 3I–K and Supplementary Tables 6 and 7).

The current study showed that suppurative AP that originated from crown caries (Fig. 4A) and periodontal inflammation, including GV, MP, and ABR, which is pathognomonic for PD (Fig. 4B), concurrently developed in periodontal tissues around carious molars in the diabetic rodents. In addition, the prevalence of periodontal inflammation was highly correlated with that of dental caries, and MP was frequently found with carious inflammation (Fig. 4C). These results suggest that carious inflammation might have a strong effect on adjacent periodontal tissue in diabetic animals. However, we could not definitively distinguish PD-derived inflammation from carious inflammation around the dental root. As a result of fluoride treatment, the cariostatic effect on teeth (14,15) in diabetic rodent models was obvious, and the progression of dental caries was markedly suppressed.

In the diabetic rodent models, fluoride treatment drastically attenuated the incidence and severity of periodontal inflammation in addition to preventing dental caries. ABR and MP persisted together with enhanced GV, and PD apparently occurred in some caries in fluoride-treated diabetic rodents. However, the lesions inevitably led directly to AP, followed by P, and were completely consistent with the incidence of dental caries. Fluoride has no inhibitory effect on oral bacteria (18), and there is no report involving the effect of the fluoride on periodontal inflammation in human and experimental animals at present. It is highly probable that fluoride could not suppress the progression of dental caries once cariogenic bacteria penetrate the tooth surface. Moreover, ABR was usually detected in diabetic animals at the apical area adjacent to the carious molars (Fig. 4C), although this lesion is characterized by a decline in the alveolar bone crest in human PD (Fig. 4B); the site of ABR completely differed between patients with diabetes with PD and diabetic rodents. Thus, these facts strongly suggest that carious inflammation may expand from the apex of a dental root and progress rostrally, thereby affecting the marginal periodontal tissue when carious inflammation penetrates the apical foramen (Fig. 4E). PD-derived inflammation might not have developed in our type 1 and type 2 diabetic animal models.

The establishment of animal models for diabetic PD is necessary because diabetes is becoming an epidemiologically important risk factor for PD in humans (1). Human PD is initiated by GV and progresses apically along the periodontal ligament, causing progressive ABR (Fig. 4B) (19). Similar PD progression is confirmed in nondiabetic rodents subjected to ligature placement, gingival inoculation with periodontal pathogens, or lipopolysaccharides (20), whereas GV surrounding noncarious molars did not progress further and cause MP and ABR associated with long-term hyperglycemia in our 2 diabetic animal models (Fig. 4D). These facts suggest that a certain threshold stimulus including above-mentioned experimental manipulation that causes the destruction of gingival tissue may also be necessary for the development of the clinical state of diabetic PD (3) (Fig. 4F). Moreover, it is certain that hyperglycemia causes dental caries in diabetic rodents (9–13,21) and that carious inflammation leads to mistakes in interpreting the development of PD in diabetic animals. Protection against dental caries is an indispensable tool for research on PD using diabetic animals, and caries inhibition by fluoride treatment is a simple but effective tool for this type of research. In conclusion, long-term hyperglycemia naturally induces dental caries but not PD in the studied type 1 and type 2 diabetic rodents.

Acknowledgments. The authors thank the Kumamoto Laboratory of the Nonclinical Research Center of LSI Medience Corporation for conducting the rat experiments. The authors also thank Nana Shako, Laboratory of Pathology, Faculty of Pharmaceutical Sciences, Setsunan University, for help with the mouse experiments.

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Author Contributions. Y.N. participated in the study conception and design, performed the experiments, analyzed data, and wrote the manuscript. K.O. participated in the study conception and design, participated in the experiments and data interpretation, and reviewed and edited the manuscript. T.M. participated in the study conception and design, participated in data interpretation, and reviewed and edited the manuscript. T.M. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.