Introduction

Systemic sclerosis (SSc) is a multi-system disorder of connective tissue characterized by excessive fibrosis in the skin and various internal organs such as the lung, kidney and heart. SSc is generally divided into two categories based on the extent of skin fibrosis: diffuse cutaneous SSc (dcSSc) and limited cutaneous SSc (lcSSc) [1, 2]. Excessive accumulation of extracellular matrix (ECM) components, especially types I and III collagen, is the most prominent pathological manifestation of the disease [3, 4]. Numerous studies have shown that an abnormally increased synthesis of these constituents in SSc skin in vivo and in cells cultured form the skin of SSc patients grown in tissue culture may involve in the development of SSc [5, 6]. The tissue fibrosis in this disorder is the main features which is caused by an excessive accumulation of ECM, especially type I collagen. It reflects an imbalance between collagen production and degradation. In many studies, fibroblasts grown from SSc patients have revealed increased synthesis of collagen [4, 7].

Matrix metalloproteinases (also named matrixin or MMPs) are the main enzymes responsible for degradation of ECM components [8]. They are essential for various normal biological processes such as embryonic development, morphogenesis, reproduction tissue resorption and remodeling [9]. MMPs also have been implicated in a number of key pathologic processes including inflammation, fibrosis, arthritis, pulmonary diseases and cancer [10]. Their activity is regulated not only at the gene expression level but is also regulated by inhibitors, including an MMP-specific family, tissue inhibitors of metalloproteinases (TIMPs). MMPs are wildly studied in SSc. Up-regulated of MMPs may implicate in the process of the disease pathogenesis, and the imbalance between MMPs and the TIMPs is proposed to be crucial in the ECM deposition in SSc. Moreover, associations of MMPs and TIMPs polymorphisms with susceptibility to SSc have been extensively reported. All these evidences indicate that MMPs may represent a novel target for the treatment of SSc.

The aim of this review is to provide an overview of the structure, function and regulation of MMPs and shortly summarize current literature on experimental findings, genetic associations of MMPs in SSc.

Structure of MMPs

The MMPs, discovered in 1962, are a family of at least 23 calcium-activated and zinc-dependent endopeptidases. They comprise a large family of protease and share several similarities in terms of their structure, regulation and function [11, 12]. More than 20 MMPs have been identified, and they are further divided into six major subfamilies based on structure and substrate specificity, including collagenases, gelatinases, stromelysins, matrilysins, membrane-type MMPs and other MMPs (Table I). The structure of all MMPs consists of three major domains: N-terminal hydrophobic signal sequence, a prodomain region and a catalytic domain [13, 14]. All MMPs except MMP-7 and MMP-22 have a C-terminal hemopexin-like domain, which determines their substrate specificity. A transmembrane domain is found in MMP-14, -15, -16, -17, -24, -25, which is important to cell-bound MMPs [15].

Table I Characteristics of MMPs
Full size table

Regulation of MMPs

The regulation of MMPs is complex and occurs at various levels, including at gene transcription, posttranslational activation of zymogens, and endogenous inhibition [16].

Many regulatory factors such as cytokines (interleukin [IL], and tumor necrosis factor [TNF]), growth factors (epidermal growth factor [EGF], transforming growth factor [TGF], and platelet-derived growth factor [PDGF]), hormones, steroids, oncogenic cellular transformations may involved in the gene expression of MMPs at the level of transcription [17, 18]. As an illustration, IL-1, IL-12, EGF, PDGF and TNF can up-regulate the expression of MMPs. However, TGF, hormones and steroids can down-regulate the expression of MMPs [19–21]. MMPs can further participate in processing and activating other MMPs. The regulatory mechanisms of MMPs induced by such stimuli may vary according to different MMPs and cell types. Furthermore, single nucleotide polymorphisms (SNPs) in gene-regulatory DNA elements may also influence MMPs transcription.

All the MMPs are synthesized and secreted in inactive forms as the latent zymogens, and are activated by the loss of a 10-kDa propeptide either extracellularly or on cell membranes. This step may occur in either intracellular or extracellular. Besides, the highly conserved cysteine residues (“cysteine switch”) in the propeptide domain have the ability to bind the zinc atom in the catalytic domain, thereby rendering the enzyme inactive [22, 23].

Activation of MMPs is regulated by TIMPs which are a group of structurally related, endogenous inhibitors. They inhibit the activity of MMPs through non-covalent binding of active forms of MMPs in the extracellular space in 1:1 molar stoichiometry [24]. The dynamic equilibrium between active MMPs and TIMPs is crucial for many diverse cellular processes including proliferation, migration, adhesion and apoptosis. Besides, TIMPs also play a critical role in maintaining the balance between ECM deposition and degradation in physiological processes [25, 26]. To date, four TIMPs have been identified: TIMP-1, -2, -3, and -4 [27]. These TIMPs are secreted by a variety of cell lines including smooth muscle cells and macrophages and their activity is increased by PDGF and TGF-β and either increased or decreased by different ILs [28].

MMPs in Inflammatory Process

Numerous studies have been reported that MMPs can either promote or inhibit inflammatory process by direct proteolytic processing of inflammatory mediators including chemokines, cytokines to activate, inactivate or antagonize their functions [29, 30]. As described in a comprehensive review on the roles of MMPs in inflammatory condition by Manicone et al. [30], inflammatory cytokines are involved in the MMPs gene expression. For example, TNF-α is expressed on T-cells and macrophages as a 26 kDa membrane-bound protein (pro-TNF-α) that is activated by cleavage to a 17 kDa soluble cytokine by TNF converting enzyme (TACE), identical to ADAM17, a member of the disintegrin family of metalloproteinases [31]. Besides, IL-1β is another potent pro-inflammatory cytokine that requires proteolytic processing before activation not only by caspase-1 but also several MMPs, including MMP-2, MMP-3 and MMP-9. Interestingly, MMP-3 can degrade the mature IL-1β cytokine, suggesting potentially dual roles for MMPs in either stimulating or inhibiting IL-1β effects [32].

MMPs and Fibrosis

Organ fibrosis is a complex process that is defined as excess deposition and accumulation of ECM in the tissues including skin, heart, lung, kidney and vessels [33]. The amount of ECM in the tissue may be controlled through balance among the ECM production, the ECM degradation by MMPs and TIMPs. Too little or too much accumulation of ECM may result in different diseases such as connective tissue disorders or organ fibrosis [34]. The decreased activity of ECM-removing MMPs is mainly due to an increased expression of their TIMPs. Evidences suggest that fibrotic livers have high expression of the TIMPs, including TIMP-1 and TIMP-2, and thus the combination of low expression of MMPs and high TIMPs may prevent the degradation of the fibrillar collagens. An expert review by Iimuro et al. [35] showed that the imbalance between too few MMPs, too much ECM and too many TIMPs possibly accounted for the advanced liver fibrosis and the failure to resolve the fibrous scar. For example, MMP-9-deficient mice exhibited significantly less pulmonary fibrosis (PF) in response to bleomycin than their with MMP-9+/+ littermates [36]. In the hepatic fibrosis model infected by Schistosoma mansoni, the severity of fibrosis was most closely associated with the increased MMP-9 activity [37]. Similarly, in response to bleomycin, mice deficient in γ-glutamyl transpeptidase showed a reduction in PF, in part associated with lower MMP-9 activity in lung tissues [38]. Besides, some researches also suggest that MMP-13 may be involve in the pathogensis of PF. In rat animal models, the expression levels of MMP-13 were decreased in lung tissue of bleomycin-induced PF in contrast to normal lung tissue [39]. Selman et al. [40] demonstrated that the expression of MMP-13 was not detected in lung tissue of people with human idiopathic PF. Asano et al. [41] suggested that MMP-13 might be involved in the fibrotic process of SSc, especially in the initiation of fibrosis.

Aberrant Expression of MMPs/TIMPs in SSc

SSc is mainly characterized by microvascular damage and excess organ fibrosis. The damage is caused by a massive deposition of collagen and other connective tissue components. In general, tissue fibrosis reflects an imbalance between collagen production and degradation. Numerous studies have shown that MMPs and TIMPs are expressed in SSc disease processes. In one study [42], patients with SSc had higher serum concentrations of MMP-9 and TIMP-1, and a higher ratio of MMP-9 to TIMP-1 than healthy controls. Serum MMP-9 concentrations were significantly higher in the diffuse type than the limited type of SSc. Serum concentrations of MMP-9 correlated well with the degree of skin involvement, as determined by the Rodnan score and with serum concentrations of TGF-β. Moreover, dermal fibroblasts from patients with SSc produced more MMP-9 than those from healthy controls when they were stimulated with IL-1β, TNF-α, or TGF-β. As we know, SSc is frequently accompanied with interstitial lung disease (ILD) often leading to lung fibrosis. ILD may develop in both lcSSc and dcSSc and is associated with decreased survival. Andersen et al. [43] showed that bronchoalveolar fluid levels of MMP-9 were enhanced in SSc patients with signs of ILD, compared to SSc patients without ILD and healthy controls. TIMP-1 levels were elevated in both patient groups compared with healthy controls. Levels of MMP-9 and TIMP-1 were inversely associated with total lung capacity, suggesting a role in the remodelling in ILD and lung fibrosis. Circulating MMP-7 levels were significantly higher in SSc patients compared with normal controls. The negatively significant correlation of MMP-7 concentrations and the diffusion capacity of the lung for carbon monoxide value together with the significant correlation with lung fibrosis and dyspnea indicated a correlation between the MMP-7 concentration and the extent of pulmonary involvement [44]. Findings from a recent study [45] on the clinical correlations of MMP-12 with SSc suggested that serum levels of MMP-12 were significantly higher in both patients with dcSSc and lcSSc, and increased serum levels of MMP-12 were associated with the extent of skin involvement, the presence of digital ulcers and severity of nailfold capillary abnormalities. Furthermore, MMP-12 was found to be strongly expressed in different cell types of SSc skin and lung tissues including fibroblasts, microvascular endothelial cells and inflammatory cells. All of these indicated MMP-12 a critical role in the development and progression of skin sclerosis, PF and peripheral vascular damage in patients with SSc. However, another study by Kikuchi K et al. [46] showed that Serum MMP-9 activity in patients with diffuse cutaneous SSc was significantly decreased compared with that of limited cutaneous SSc or normal controls. Asano et al. [41] found that the serum levels of MMP-13 were significantly decreased in SSc patients, suggesting that the decreased expression of MMP-13 contributed to the establishment of fibrosis through reducing the degradation of ECM proteins.

Genetic Polymorphisms of MMPs and SSc

It has now been established that sequence variation in the promoter regions of MMP genes may alter the level of gene expression and thus influence MMP/TIMP balance within the ECM. The association of MMPs gene SNPs with SSc has been extensively studied, but the controversial results exist. Marasini et al. [47] showed that individuals who carried 6A allele of MMP-3 might be relevant for susceptibility to and pathogenesis of SSc. For MMP-12 promoter region, the SNP rs2276109, has been studies in Italian population with SSc [48]. A significant difference was observed in MMP-12 rs2276109 genotype distribution between patients with SSc and controls (P = 0.0003), and between lcSSc and dcSSc (P = 0.003). This gene polymorphism might contribute to susceptibility to SSc, and in particular to dcSSc and PF. In a large cohort of European Caucasian, Wipff et al. [49] investigated six SNPs (rs-1306, rs-790, rs-735, rs1292301, rs7201, rs243849) of the MMP-2 gene, two SNPs (rs17576, rs2274756) of MMP-9 and two SNPs (rs743257, rs1042703) of MMP-14 genes with the susceptibility to SSc. They did not find any association between the investigated polymorphisms and SSc susceptibility nor with the major SSc clinical subsets. Skarmoutsou et al. [50] showed that the presence of the polymorphism −1562 C/T (rs3918242) of MMP-9 was not associated with the susceptibility to SSc in the Italian population, but it seemed to suggest that the C/T genotype of the MMP-9 −1562 C/T polymorphism might be associated with a smaller risk for susceptibility to developing skin ulcers only in male SSc patients. A same polymorphism in the MMP-1 promoter resulting from a guanine insertion at −1607 bp had been identified in different populations. Johnson et al. [51] failed to detect any association with the development and clinical manifestations of SSc in Caucasians, African-Americans, and Hispanics. Similar results were also found on this issue, but the high activity promoter genotype of MMP-1 seemed to significantly correlate with the limited subset of Korean SSc patients [52]. Besides, Indelicato et al. [53] analyzed three polymorphisms of the TIMP-1 gene (−19A/G, +261 C/T and +372 T/C) in Italian population. Their findings suggested that the +372 T/C polymorphism was associated with SSc in male individuals, but no association with the clinical characteristics of SSc Italian patients and TIMP-1 gene polymorphisms was observed.

Given the complexity of the mechanisms driving SSc and the heterogeneity of the studied populations, further studies will be necessary. However, upon validation and confirmatory mechanistic studies, MMP polymorphisms may be a valuable tool for clinical stratification of SSc patients.

Therapeutic Value of MMPs in SSc

As described above, the discovery of MMPs may be a new field of investigation for the development of new therapeutics in the treatment of SSc. Sato et al. [54] showed that anti-MMP-1 autoantibody contributed to the development of fibrosis by inhibiting MMP-1 collagenase activity and reducing the ECM turnover, and anti-MMP-1 autoantibody might also contribute to SSc vasculopathy through inhibition of angiogenesis as well as vascular damage due to fibrosis. As a result, they pointed out that MMP inhibitors, including anti-MMP-1 antibody, would be potential therapeutic targets for SSc. Nishijima et al. [55] also found similar results, and suggested that anti-MMP-3 antibody might be related to various aspects of SSc disease expression, including skin sclerosis, lung fibrosis, renal vasculopathy and autoimmunity. Apart from these, Serrati et al. [56] suggested that MMP-12 over-production in SSc fibroblasts and SSc endothelial cells might have a critical pathogenic role in SSc-associated vascular alterations, and MMP-12 could be therapeutically effective in SSc by blocking angiogenesis. In addition, over-production of MMP-12 by SSc microvascular endothelial cells might contribute to reduced angiogenesis in SSc patients, suggesting clinical relevance in therapeutic attempts to reverse the process of anti-angiogenesis [57, 58].

Conclusion

SSc has the highest case fatality rate among the connective tissue diseases. Although clinical outcomes have improved in recent years, no current therapy is able to reverse or slow the natural progression of this disease [59]. Thus, SSc is considered incurable, and so determining the mechanisms underlying the disease is a priority for research efforts. As all we known, MMPs play an important role in both physiological and pathological condition. The degree of this involvement can be modulated either by increasing or decreasing MMP expression and activity. Their role in SSc is unfolding as we obtain more information implicating their presence in inflammatory and fibrosis process. We propose that MMPs may play an important role in the etiology of SSc. In spite of the presence of few conflicting data regarding MMPs in SSc, generally this enzyme can be regarded as participating in crucial steps in the pathogenesis of autoimmune diseases. Therefore, it should be considered a target for therapy in SSc.