Periodontal disease is a complex inflammatory disease initiated by pathogenic plaque biofilms and results in destruction of tooth-supporting tissues and alveolar bone [17, 18]. Proteolytic enzymes like MMPs play a major role in the degradation of collagens in periodontal tissues. The expression and regulation of MMPs and TIMPs in HGFs are therefore crucial for maintenance of tissue homeostasis and periodontal health. Although many studies have been performed to elucidate the mechanisms involved in the synthesis and regulation of MMPs in periodontal research, no studies are available on the effect of P. gingivalis LPS structural heterogeneity on the expression of MMPs and the underlying regulatory mechanisms.
MMP-3 is known as stromelysin which has both elastinolytic and collagenolytic activities that degrade basement membrane components such as laminin, elastin fibronectin as well as collagen types II, III, IV, V, IX, X and XI [8, 19]. Its level could significantly increase following the stimuli of pro-inflammatory cytokines, growth factors and LPS [14, 20–22]. It has been shown that HGFs could upregulate the expression of MMP-3 due to the effects of pro-inflammatory cytokines such as IL-1β and TNF-α [23–25]. The current study showed that the expression of MMP-3 mRNA and protein was markedly upregulated by P. gingivalis LPS1690, whereas no induction was observed in cells treated with P. gingivalis LPS1435/1449, indicating that the heterogeneous lipid A structures of P. gingivalis LPS may differentially modulate the expression of MMP-3 in HGFs. Moreover, TIMP-1 expression was differently modulated by the two isoforms of P. gingivalis LPS as well. It functions as an inhibitor of MMPs by forming non-covalent complexes with MMPs. It has recently been shown that MMP-3 and TIMP-1 variants may significantly contribute to chronic periodontitis and disease progression . The imbalance between MMPs and TIMPs has been implicated in periodontal tissue destruction .
P. gingivalis has long been recognized as a major periodontopathogen . Recently, it is regarded as a keystone pathogen due to its ability to significantly influence the oral microbial community by modulating the innate host response [29, 30]. Moreover, this bacterium adopts multiple pathogenic mechanisms to evade or subvert the host immune system [31–33]. Notably, P. gingivalis LPS exhibits significant structural heterogeneity with both isoforms of LPS1435/1449 and LPS1690, and our recent studies show that they differentially affect the innate host defense and underlying signaling pathways, thereby contributing to the pathogenesis of periodontal disease [4, 34, 35]. The current observation that the different isoforms of P. gingivalis LPS modulate the expression of MMP-3 and TIMP-1 may represent an additional pathogenic mechanism adopted by this noxious species to disturb the physiological tissue remodeling and tissue homeostasis, leading to the initiation of periodontal disease.
P. gingivalis and its virulence attributes such as LPS can stimulate various cells types to secrete MMPs including MMP-3 [36, 37]. On the contrary, some studies have suggested that P. gingivalis LPS may not induce MMPs such as MMP-1, -2 and −9 . A study performed on gingival epithelial cells using P. gingivalis LPS and E. coli LPS showed that neither LPS nor IL-1β induced MMP-2 or MMP-9 . Studies on tissue models such as synovial membranes dissected from rat knee joints showed induction of MMP-1, -3 and −9 mRNA levels but not MMP-2 in response to LPS stimulation . However, foregoing studies have not considered the heterogeneous nature of bacterial LPS lipid A structures. Therefore, the conflicting findings of the previous studies could to some extent be due to different isoforms of P. gingivalis LPS as demonstrated in the present study.
In the present study, E. coli LPS-treated HGFs exhibited rapid and significant induction of MMPs 1 and 2 mRNAs with reference to the cells treated with P. gingivalis LPS1690. One possibility for this observation may be the higher responsiveness of HGFs to hexa-acylated nature of the E. coli LPS as compared to the penta-acylated structure of P. gingivalis LPS1690. This notion is consistent with previous findings that E. coli LPS is a potent inducer of the production of MMPs in fibroblast-like synovial cells and rat chondrocytes, as well as other innate host response molecules in HGFs and gingival/oral epithelia [41, 42]. Moreover, it was noted that both P. gingivalis LPS1435/1449 and E. coli LPS significantly upregulated the expression of MMP-2 mRNA but not its protein as compared to the controls. A number of factors may account for this finding, such as the stability of mRNA, its processing and splicing patterns, half-life of the target protein and post-translational modifications [43, 44]. Therefore, in the present study increase in MMP-2 mRNA expression level may not be necessarily reflected at its protein level.
TIMPs exhibit high affinity for binding with MMPs and lead to inhibition of their activities. In the present study, TIMP-1 mRNA was upregulated by P. gingivalis LPS1435/1449-treated HGFs, while no significant up-regulation was observed in P. gingivalis LPS1690-stimulated cells. The current results may not be comparable with previous studies in which the structural heterogeneity of LPS was not fully considered [45–49]. This omission may account for the conflicting reports in the literature. Hence, some studies have observed lower TIMP-1 levels in the conditioned media of HGFs in response to P. gingivalis LPS . In contrast, other studies have noted the increased expression level of TIMP-1 in gingival crevicular fluid of periodontitis patients [45, 47]. Moreover, periodontal treatment could alter the balance between MMP-3 and TIMP-1 [46, 48]. Based upon the current findings, further study may be warranted to explore the association of different isoforms of P. gingivalis LPS with periodontal conditions in periodontal patients and the possible effect of periodontal treatment on the expression of these LPS isoforms by P. gingivalis. In addition, the discrepancy observed in TIMP-1 mRNA and protein expression following the stimulation of both P. gingivalis LPS1435/1449 and E. coli LPS in HGFs could be due to the complex regulation of transcription and translation [43, 44].
LPS is the major immuno-stimulatory component of P. gingivalis which has shown to be capable of interacting with TLRs. Binding of LPS to TLRs activates the downstream signal transduction pathways such as NF-ĸB and MAPK [50, 51]. Previous studies have suggested that the activation of MMPs could be through both NF-ĸB and MAPK signaling [23, 52–54]. The present study demonstrated that p38 MAPK and ERK are critically involved in P. gingivalis LPS1690- and E. coli LPS-induced expression of MMP-3 in HGFs. This finding is supported by a previous study that p38 MAPK and ERK1/2 pathways are essential for the expression and regulation of MMPs in various cell types in response to LPS . ERK, JNK and p38 MAPK pathways play vital roles in regulating the expression of MMPs induced by various stimulants such as cytokines [53, 55, 56]. It is noteworthy that the nature of the stimuli could result in specific signal transduction pathway in the same cell type. For instance, MAPK inhibitor significantly reduced the MMP-3 production in HGFs stimulated with IL-1β, but not with epidermal growth factor . In addition, NF-ĸB pathway may be involved in regulation of MMP-3 expression in rabbit dermal fibroblasts, human saphenous vein and rabbit aortic smooth muscle cells [57, 58]. The present study showed that NF-ĸB signaling is not critically involved in LPS-induced MMP-3 expression in HGFs. Notably, the MAPK pathway but not NF-κB was significantly involved in the regulation of MMP-3 expression in HGFs in both mRNA and protein levels. Previous studies have also proven that the expression of MMP-3 is mainly mediated through P38 MAPK, ERK and tyrosine kinase pathways, but not through NF-κB pathway [23, 59, 60]. Moreover, although a study reported that the activation of NF-κB could be important for MMP-3 secretion, no consensus NF-κB binding site was identified in the MMP-3 gene promoter [61, 62]. It suggests that NF-κB may regulate the expression of this gene through different binding sites or interacting with other transcription factors . Therefore, within the context and limitations of the present study, it is tempting to speculate that MAPK pathway may be crucial for MMP-3 expression in HGFs in response to P. gingivalis LPS1690. Furthermore, it would be interesting to extend the study to other cells types in human gingiva like gingival epithelial cells to ascertain whether MAPK pathway plays a predominant role in the expression and regulation of MMP-3 in other cells of oral tissues.