Matrix metalloproteinases (MMPs) are a family of enzymes which, in concert, are capable of degrading collagen. We investigated whether human MMPs could participate in the degradation of dentin organic matrix after demineralization. We performed Western blot analyses using MMP-specific antibodies to identify MMPs in human dental caries lesions. Enzymography and functional activity assays, with 125I-labeled gelatin as substrate or quantitating the degradation of type I collagen, were used to determine the activity of purified and salivary gelatinolytic (MMP-2 and MMP-9) and collagenolytic (MMP-8) enzymes with and without acid-activation in pHs relevant to caries. Respective analyses were done with caries-related bacteria. We performed electron microscope analyses to assess the degradative activity of sterilized salivary host MMPs on demineralized human dentin. Human MMP-2, MMP-8, and MMP-9 were identified in demineralized dentinal lesions. The latent purified forms of these enzymes were activated at low pH (4.5), followed by neutralization, mimicking the conditions during caries progression. Incubation of human saliva at low pH followed by neutralization resulted in a four-fold increase in the gelatinolytic activity. No gelatinolytic or collagenolytic activity was observed in bacterial samples. The activated enzymes in saliva degraded demineralized dentin organic matrix in vitro. These results demonstrate the pH-dependent activation mechanism of MMPs, which may have a distinct role in different physiological and pathological conditions. They further demonstrate that host MMPs, activated by bacterial acids, have a crucial role in the destruction of dentin by caries.

No abstract

Previous studies suggest that salivary and pulp-derived host enzymes, matrix metalloproteinases (MMPs), may be involved in dentin caries pathogenesis. To study the inhibition of acid-activated human salivary MMPs by non-antimicrobial chemically modified tetracyclines (CMTs), we used a functional activity assay with 125I-labeled gelatin as a substrate. To address the role of MMPs in the progression of fissure caries in vivo, we administered the MMP inhibitors CMT-3 and zoledronate to young rats per os for 7 weeks, 5 days a week. Caries lesions were visualized by Schiff reagent in sagittally sectioned mandibular molars. Marked reduction in gelatinolytic activity of human salivary MMPs was observed with CMT-3. CMT-3 and zoledronate, both alone and in combination, also reduced dentin caries progression in the rats. These results suggest that MMPs have an important role in dentin caries pathogenesis, and that MMP inhibitors may prove to be useful in the prevention of caries progression.

MMP-20 (enamelysin), the matrix metalloproteinase family member discovered in the enamel organ, has also been detected in odontoblasts during dentin formation. We studied the presence and localization of MMP-20 in mature human teeth in health and disease. In immunohistochemistry, MMP-20-positive staining was observed most intensively in the radicular odontoblastic layer and also in dilated dentinal tubuli of caries lesions. By Western blotting, MMP-20 was detected in odontoblasts and pulp tissue of both sound and carious teeth, in dentinal fluid and dentin of sound teeth, but not in soft carious dentin. We conclude that MMP-20 produced during primary dentinogenesis is incorporated into dentin and may be released during caries progression. The main cellular source of MMP-20 in the dentin-pulp complex is the odontoblasts, which secrete MMP-20 into the dentinal fluid.

Previous studies have demonstrated that (at least) matrix metalloproteinase (MMP)-2, -8, -9, -14 and -20 are expressed by human odontoblasts. Here, we analysed the expression of 19 MMPs and their specific tissue inhibitors (TIMP)-1, -2 and -3) -1, -2 and -3 in mature human odontoblasts and pulp tissue. Since MMP-20 is almost exclusively expressed by the dentin-pulp complex cells, we further analysed the effect of transforming growth factor (TGF)-beta1 and bone morphogenetic protein (BMPs)-2 on its expression. Matrix metalloproteinase-9 served as a positive control for growth factor responsiveness. It was found that MMP-1, -2, -9, -10, -11, -13, -14, -15, -16, -17, -19, -20 and -23, in addition to TIMP-1, -2 and -3 were expressed by both odontoblasts and pulp tissue. Neither MMP-3 nor MMP-12 were expressed in odontoblasts or pulp tissue, and MMP-7, -8, -24 and -25 were expressed only in the odontoblasts; MMP-2, -10, -11, -14 and -20 were expressed more abundantly by odontoblasts, whereas pulp tissue expressed more MMP-13 and MMP-17. Transforming growth factor-beta1 (1 ng ml(-1)) and BMP-2 (100 ng ml(-1)) did not markedly affect MMP-20 mRNA expression. In contrast, TGF-beta1 alone and with BMP-2 significantly upregulated MMP-9 mRNA by 2.4-fold and by 2.6-fold, respectively, in odontoblasts, while in pulp tissue no effects could be detected. The wide-scale expression of MMPs and TIMPs by mature human odontoblasts and pulp tissue suggests that they may participate in dentin matrix organization prior to mineralization, and that growth factors may further control dentin matrix modeling by differentially regulating individual MMPs.

Recent findings show that matrix metalloproteinase-8 (MMP-8) is expressed, in addition to neutrophils, by human chondrocytes, cultured fibroblasts, and endothelial cells. We investigated the expression of MMP-8 in other human mesenchyme-derived cells, odontoblasts, and pulp tissue. Odontoblasts and pulp tissue were collected from extracted human teeth for MMP-8 mRNA analysis with reverse-transcription/polymerase chain-reaction (RT-PCR) and Southern blot. The expression, localization, and secretion of MMP-8 protein were studied with Western blot, immunohistochemistry, and immunofluorometric assay. The effect of TGF-beta1 (10 ng/mL) on the expression, secretion, and concentration of secreted MMP-8 was studied by odontoblast and pulp tissue culture methods (Tjäderhane et al., 1998a). RT-PCR demonstrated MMP-8 mRNA expression in native and cultured odontoblasts and pulp tissue and cultured pulp fibroblasts, with a 522-bp transcript comparable with that of bone marrow cells. The specificity of PCR was confirmed with Southern blot. Western blot with MMP-8-specific antibody detected 65- and 50-kDa proteins in native samples, representing latent and active forms of mesenchymal-type MMP-8, and in the conditioned odontoblast culture media, 50-kDa protein was observed. TGF-beta down-regulated the MMP-8 mRNA and concentration of secreted protein in both cultures. Immunohistochemical staining detected MMP-8 in odontoblasts. These findings indicate that mesenchyme-derived cells of the dentin-pulp complex express, synthesize, and activate MMP-8, which may, in concert with odontoblast-derived gelatinases, participate in organization of dentin organic matrix prior to mineralization.

Odontoblasts cannot be cultured by traditional cell culture methods, thus restricting in vitro studies. Here we present an organ culture method for human odonto-blasts that utilizes the pulp chamber as a culture crucible. Crowns of human third molars were dissected, pulp was gently removed, and the odontoblasts attached to and in the walls of the pulp chambers were cultured in serum-free OPTI-MEM medium, or DMEM/Ham's F12 medium containing 10% serum. Pulp tissues were cultured separately. Cell content and morphology were analyzed by SEM, and the removed pulps were examined by light microscopy. Proteins secreted into the medium with or without TGF-beta1 supplementation were metabolically labeled with [35S]methionine, and the total protein content was assessed by TCA precipitation and SDS-PAGE/fluorography. To assess the role of gelatinolytic enzymes on dentin matrix remodeling, we used enzymography to analyze the effect of TGF-beta1 on gelatinase A and B expression. SEM revealed odontoblasts in pulp chambers after 5 days of culture, with only few or no fibroblasts, and no alterations in the odontoblast cell morphology or differences between the cells cultured in serum-free and serum-containing media. Rarely were any odontoblasts present in pulp tissue. Radiolabeling revealed protein synthesis and secretion until day 6 in both the odontoblast and pulp cultures, with no marked differences between TGF-beta1-treated and control cultures. The level of gelatinase A remained constant up to 7 days, while gelatinase B expression was always low and decreased with time in culture. However, gelatinase B levels were markedly increased upon TGF-beta1 treatment of cells and remained high to day 7. The results suggest that this method provides a novel technique for the study of human odontoblasts in vitro and that odontoblasts can be cultured even in serum-free conditions.

This study demonstrates that the presence of yeast, but not lactobacillus infection, is related to salivary flow rate. Salivary pH is correlated with the primary infection of both lactobacilli and yeasts. The low output of saliva appears to influence the quantity of lactobacilli more than that of yeasts. The flow rate was lower and the presence of yeasts was higher in females than in males.