| Matrix metalloproteinases (MMPs) and their specific tissue inhibitors (TIMPs) in mature human odontoblasts and pulp tissue | ||
|---|---|---|
| Prev | Chapter 6. Discussion | Next |
This thesis shows that mature human odontoblasts and pulp tissue are capable of expressing various MMPs and TIMPs, in addition to those observed earlier: MMP-2, MMP-9 and MMP-20. Furthermore, TGF-β 1 and BMP-2 differentially regulate the expression of these proteins. It is still unknown why so many members of different subclasses of metalloproteinases are synthesized in dentin-pulp complex cells, and what their functions might be in human dentin-pulp complex. There are only a few studies demonstrating the role or function of an individual MMP in the dentin-pulp complex in vivo in physiological or pathological conditions (Hall et al. 1999, Wahlgren et al. 2002).
MMPs may have dual-roles; they can act in both healthy and disease conditions. Thus, the possible roles of MMPs in mature odontoblasts could be classified in the following subcategories: 1) Functions of MMPs in intact and healthy teeth in processes of physiological secondary dentin formation and mineralization, 2) MMPs participating in matrix degradation during dental injury, 3) Roles of MMPs in tertiary dentinogenesis, and 4) Roles of MMPs in pulpal inflammation.
MMPs are suggested to play a role in the initiation of cartilage mineralization (D’Angelo et al. 2001). During dentin matrix mineralization, MMPs may take part in collagen matrix organization, before mineral deposition occurs. During the organization of collagen fibrils other proteins such as proteoglycans are involved (Jones & Boude 1984, Boskey 1989), and if they need to be cleaved prior to mineralization, it is most probably done by MMPs. There is evidence that hydroxyapatite crystals inactivate and induce autocatalytic degradation of both MMP-1 and MMP-3 in vitro (Kremer et al. 1998). Therefore, it is possible that the MMP activity during predentin matrix degradation is at least partially regulated by the formation of hydroxyapatite crystals at the mineralization front.
So far, there is only one histological analysis showing that MMP-3 is located in the predentin, and thus, it is suggested to participate to the regulation of dentin mineralization by regulating the presence and activity of different proteoglycans in the different parts of predentin (Hall et al. 1999). However, this work is done with continuously growing rat incisors, and since we did not detect MMP-3 in mature human odontoblasts, it is possible that in humans another member of stromelysin family, MMP-10 (stromelysin-2), is responsible for the function suggested for MMP-3 in rat incisors, since relatively high expression levels were detected for this MMP in the odontoblasts (Table 2). Therefore, it is possible that significant differences occur in the MMP expression profiles in dentin-pulp complex between species.
If a tooth is affected by an injury and dentin demineralization occurs, which leads to a perturbation of tooth homeostasis, and possibly to uncontrolled activities of MMPs, it is highly possible that TGF-β 1 may participate in the regulation of different dentin-pulp complex defensive reactions. TGF-β 1 may, for example, protect the pulp tissue from degradation in a similar way that it protects cartilage collagen from destruction, by reducing collagenases (Hui et al. 2000). Since glycosaminoglycans and decorin, which both exist in dentin, are able to bind growth factors (Yoshiba et al. 1996, Imai et al. 1997), and latent TGF-β is stored in the ECM (Dallas et al. 1995), it is possible that MMPs bound to dentin matrix serve as a reservoir of enzymes capable of releasing matrix bound latent TGF-β 1 and activating it (Allan et al. 1991, 1995, Imai et al. 1997, Martin-De Las Heras et al. 2000, Yu & Stamenkovic 2000). Thus MMPs may be involved in the TGF-β mediated cellular responses occuring during dental injury. However, as a response to injury, TGF-β may not stimulate type I collagen synthesis (I), but may indirectly affect dentin matrix synthesis by regulating, alone or synergistically with BMP-2 and possibly other dentin-bound growth factors, the expression of MMPs. Indeed, differential effects on several MMPs for both growth factors were detected in these (II, III, III) and in previous studies (Tjäderhane et al.1998b). During prolonged dental injury, it is highly probable that TGF-β may induce BMP-2 expression, as has been detected with pulpal cells (Calland et al. 1997). Together they may have a positive role in promotion of pulpal healing, and this effect may be at least partially mediated by the regulation of MMP expression and synthesis.
If the dental injury is severe and leads to odontoblast cell death, TGF-β may induce new odontoblast-like cell differentiation from pulpal stem cells and their migration to the pulp chamber edge, where they are responsible for dental collagen matrix synthesis and mineralization (D’Souza et al. 1995, Melin et al. 2000). In the present study, TGF-β 1 and BMP-2 seem to only induce the expressions of pulpal MMPs of MMP-1, -2, -13 and -17, of which MMP-13 and MMP-17 are expressed at much higher levels by pulp tissue compared to odontoblasts. Since the expression level of MMP-8 was quite low in the pulp tissue compared to MMP-13 (IV), it is possible that MMP-13 is the main collagenase in the pulp tissue together with MMP-1. This is supported by the recent study utilizing cDNA microarray, in which MMP-13 exprssion was extremely high in the pulp tissue compared to all the other MMPs studied (Sulkala et al. submitted for publication). Although the steady state expression of MMP-1 is low in the healthy tooth (IV), growth factors induced the expression of MMP-1. Since pulp tissue inflammation is an essential defensive reaction in the protection of dentin-pulp complex, and since growth factors mainly induced the expression of MMPs, it is possible that MMPs may have a role during pulpal inflammation, by mediating the breakdown of the pulpal connective tissue enabling migration of the odontoblast-like cells, as has been suggested for MMP-2 in the induction of epithelial cell migration (Giannelli et al. 1997).
TIMPs have widespread biological functions in addition to MMP-inhibition, e.g. they may stimulate cell growth (Hayakawa et al. 1992, Nemeth et al. 1996). Since the present study shows that TIMP-2 is not regulated by TGF-β and BMP-2 (see Fig. 14, Table 2), it is possible that TIMP-2 is constitutively expressed in the dentin-pulp complex, and one of its physiological functions may be activation of MMPs, as has been indicated with MMP-2 (Wang et al. 2000). Whether TIMP-2 is also included in the MT1-MMP mediated proMMP-20 activation remains to be studied. TIMP-1 is able to bind collagen fibrils bound active MMP-1 (Allan et al. 1991), and since TGF-β 1 and BMP-2 induce the expression of TIMP-1 in human odontoblasts, it is highly possible that TIMP-1 may indirectly inhibit collagen matrix destruction by inhibiting active MMP-1. Furthermore, since type III collagen is the main substrate of MMP-1, and type III collagen is a component of a reparative dentin, it is possible that as a response to exernal irritation, type III collagen in the predentin is no longer degraded by MMP-1, allowing dentin mineralization to proceed to reparative dentin.
Of the TIMPs, TIMP-3 has the ability to bind to the ECM, especially to matrix components containing glycosaminoglycan chains, such as heparin, and sulphated components (Langton et al. 1998, Yu et al. 2000). TIMP-3 may serve as a reservoir of inhibitors in the dental matrix, and during dental injury it may be released from the matrix to control activities of MMPs. It is possible that TGF-β is released from the matrix during dental injury, and since we show that TGF-β 1 in the pulp tissue alone, and together with BMP-2 in odontoblasts induces TIMP-3, it is possible that TIMP-3 released from the matrix inhibits the activities of MMPs during dental injury. In addition, TIMP-3 may also induce cell apoptosis (Bond et al. 2000). The possible role of TIMP-3 in apoptosis of odontoblasts as a response to severe caries would be an interesting hypothesis to study.
Contrary to other pulpal TIMPs, the present study shows that TIMP-4 expression is inhibited by TGF-β 1 and BMP-2. Since TIMP-4 is a general inhibitor of many MMPs (Stratman et al. 2001), it is possible that during dental injury inactivation of TIMP-4 permits the possible MMP mediated matrix degradation during pulpal cell migration, indicating that TIMP-4 may have an indirect healing effect on dentin matrix as a response to injury. For example, TIMP-4 regulates the activity of MMP-2 either directly or through inhibition of MT1-MMP (Bigg et al. 2001, Hernandez-Barrantes et al. 2001). Since the growth factor down-regulation of TIMP-4 seemed to be accompanied by the induced expression of MMP-2 (Table 2), TIMP-4 may have a role in the increased activity of MMP-2 in pulp tissue (Table 2).
Even though the best-known effect of MMPs and TIMPs is the degradation of proteins either in morphogenesis and growth of tissue or associated with tissue catabolic activities, their role is not limited to protein degradation as such (reviewed in Sternlicht & Werb 2001, Lopez-Otin & Overall 2002). It has been increasingly appreciated that by highly specifically recognizing and hydrolysing their substrates, proteinases are important regulators of diverse biochemical reactions both within the cell and in the ECM. Proteolysis is more than simple degradation; it is an important mechanism to regulate activities of proteins, for example by controlling their shedding from the cell, and their localization. Furthermore, proteinases may activate or inactivate other proteinases, enzymes, growth factors and cytokines, and hence they have central role in diverse biological events such as in cell proliferation, differentiation and migration, and tissue remodelling and apoptosis (reviewed in Lopez-Otin & Overall 2002). There is also increasing evidence that TIMPs also have roles other than regulation of MMP-activity. They have, for example, growth factor-like activity and can inhibit angiogenesis (reviewed in Gomez et al. 1997). Therefore, the roles of individual MMPs and TIMPs, either alone or acting in different combinations, may prove to be much more diverse than what we know today in many tissues, including the dentin-pulp complex.