| Matrix metalloproteinases (MMPs) and their specific tissue inhibitors (TIMPs) in mature human odontoblasts and pulp tissue | ||
|---|---|---|
| Prev | Next | |
Type I collagen is a fundamental component of a fibrous organic framework of dentin. However, there is evidence that the dentin organic matrix also contains other collagens. Collagens are triple helix molecules of either homo- or heteropolymers, but only proα2(I) (Lukinmaa et al. 1992) and proα1(III) (Lukinmaa et al. 1993) collagen mRNA have been shown to be expressed by human odontoblasts. Even though some animal studies suggest that type III collagen is not synthesized by odontoblasts (Munksgaard et al. 1978, 1980, d’Souza et al. 1995), others have detected type III collagen (Karjalainen et al. 1986) or procollagen (Becker et al. 1986) in human predentin and odontoblasts, in the dentin of dentinogenesis imperfecta-patients (Karjalainen & Söderling 1984, Waltimo et al. 1994) and in human reparative dentin (Magloire et al. 1988b). Also type VI collagen has been observed in intact dentin (Becker et al. 1986).
Even though odontoblast secretion has been investigated for years, surprisingly little is known about the expression and regulation of the collagens in human odontoblasts and pulp tissue, since most of the studies have been performed with animal models. The development of an organ culture method for mature human dentin-pulp complex tissues has enabled investigation of the secretory activity of both mature human odontoblasts and pulp tissue (Tjäderhane et al. 1998b). With this method, mRNA of proα1(I) chain of type I collagen, and type I procollagen were found to be expressed by odontoblasts and pulp tissue, and that odontoblasts abundantly synthesized type I collagen protein compared to the respective amount of pulp tissue (I), confirming previous findings (Lukinmaa et al. 1992, 1993). Furthermore, this work showed for the first time that mature human odontoblasts indeed synthesize and secrete type III collagen protein (I). The expression of type III collagen in pulp tissue was not analyzed, since it has been well characterized that pulp tissue synthesizes type III collagen (van Amerongen et al. 1983).
One reason for the apparent discrepancy between these results and some early animal experiments (Munksgaard et al. 1978, 1980) may arise from differences in protein composition between different species (Linde & Goldberg 1993). Alternatively, in the earlier studies, the extensive purification processes might have caused the loss of minor collagenous components (Karjalainen et al. 1986). For example, pepsin treatment has been shown to lead to marked underestimation of type III collagen levels in other tissues (Burke et al. 1977). The reason for type III collagen expression by mature intact human odontoblasts is unclear, but it may be needed for organization of the dentin organic matrix prior to mineralization, since type III collagen is a regular component of soft tissue ECM, but is not present in normal mineralized human tissues, including bone and dentin (e.g. Becker et al. 1986). The virtual absence of type III collagen in physiological human dentin suggests that the protein is degraded and/or removed from the predentin during the organization of the matrix. In pathological situations, this degradation and removal may be hindered and type III collagen may be left in the predentin and mineralized dentin, possibly because of alterations in MMP expression. Type III collagen may, in turn, influence the structure of dental tissues (Lukinmaa et al. 1993).
TGF-β 1 has been suggested to increase protein synthesis, including type I collagen, as a response to dental injury (Smith et al. 1995, Sloan & Smith 1999), but the precise mechanisms of how TGF-β 1 regulates the responses to injury are still not well known. For example, Sloan and Smith (1999) suggest that the increase in the predentin width after TGF-β treatment is due to increased odontoblast matrix synthesis. However, TGF-β has been shown to reduce the expression of proteins involved in dentin matrix mineralization, such as dentin sialophosphoprotein and alkaline phosphatase (Ibbotson et al. 1989, Shibata et al. 1993, Nakashima 1992, Shirakawa et al. 1994). Therefore, the predentin widening observed in TGF-β 1 treated rodent incisor slices may be caused by the reduced organic matrix mineralization. This assumption is supported by in vitro studies with the human odontoblast culture method showing that TGF-β may not induce dentin formation by up-regulating collagen synthesis, since TGF-β 1 had no effect on either type I or type III collagen levels in mature human odontoblasts (I). These findings are further supported by a study from Shibata and colleagues (1993), showing that collagen synthesis levels in osteoblasts are affected by the state of cell differentiation, and in mature osteoblasts TGF-β 1 does not affect collagen synthesis. Alternatively, these seemingly conflicting findings may again reflect the differences between species that have been demonstrated in several other cases (Linde & Goldberg 1993).
On the other hand, the data shows that TGF-β 1 up-regulates proα1(I) collagen mRNA in other cell types such as gingival fibroblasts (I), which is in accordance with other studies (Wrana et al. 1986, James et al. 1998). Pulpal fibroblasts responded variably to a TGF-β stimulus, and the effect was highly dependent on culture conditions (I), as has been previously suggested (Moule et al. 1995, Stanislawski et al. 1997). In addition, the expression of proα1(I) collagen level in pulpal fibroblasts was different from that of pulp tissue after TGF-β treatment in serum free conditions (I).
Taken together, the present data suggests that TGF-β has no direct effect on the synthesis of collagens in mature human odontoblasts. However, TGF-β may still indirectly affect dentin matrix formation and mineralization by either regulating the differentiation of replacement odontoblasts or by regulating the expresson of other proteins such as matrix metalloproteinases. The conflicting results with the pulp fibroblast cultures under different culture conditions confirm the previous suggestion that the traditional cell cultures with the pulpal cells do not necessarily provide results that can be interpreted as the same as an in vivo situation (Moule et al. 1995, Stanislawski et al. 1997). Organ culture models, such as used here (I–IV), or the human tooth slice culture-technique (Melin et al. 2000), provide important tools to investigate the responses of different tissue types to various stimuli, and to compare results between different species.