Article Text
Abstract
The metastasis-associated protein S100A4 belongs to the large family of S100 calcium-binding proteins that appear to play regulatory roles in diverse biological activities. Moreover, a prognostic role of S100A4 has been suggested for patients with several types of cancer. Cancer promoting properties for S100A4 have been demonstrated, particularly through its regulation of cell motility, proliferation and apoptosis, as well as by stimulation of angiogenesis and remodelling of the extracellular matrix.
Increased expression of S100A4 mRNA has been detected in proliferating synovial fibroblasts in rheumatoid arthritis. Furthermore, strong upregulation of the S100A4 protein in rheumatoid arthritis synovial tissue compared with osteoarthritis and control tissues has been demonstrated recently, especially at sites of joint invasion. Several immune and vascular cells were also identified to be producing S100A4 within the synovium. The local upregulation of S100A4 was accompanied by high plasma and synovial fluid concentrations of the S100A4 protein existing in the bioactive oligomeric form in patients with rheumatoid arthritis. Consistent with data from cancer studies, the extracellular S100A4 oligomer appears to be involved in regulation of several matrix-degrading enzymes and modulation of the transcriptional activation function of the tumour suppressor protein p53 in rheumatoid arthritis synovial fibroblasts.
Taken together, one can speculate that increased S100A4 protein in circulation and locally at sites of inflammation, particularly at sites of joint destruction, might be linked to the process of aggressive fibroblast behaviour contributing to the pathogenesis of chronic autoinflammatory diseases such as rheumatoid arthritis.
Statistics from Altmetric.com
The S100 proteins are small acidic calcium binding proteins (10–12 kDa) that are found exclusively in vertebrates. The name of the whole group “S100” relates to its solubility in 100% ammonium sulfate solution.1 This group consists of 20 members and is considered to be the largest subgroup of the EF-hand calcium binding protein family. S100 proteins share some structural similarities with the well known calcium-binding protein calmodulin and carry two calcium-binding EF-hand motifs with different affinities to bind calcium.2 Binding of calcium results in conformational changes to the proteins, leading to exposure of the hydrophobic surface that enables the S100 proteins to interact with a variety of target molecules.3 However the biological activity of some S100 proteins does not require calcium because they can also be regulated by zinc and copper.4
S100 proteins were first identified in 1965 by Moore in bovine brain.1 Later it was shown that their biological activity appears to be dependent on the ability to form dimers.5 In the intracellular environment, most S100 proteins exist as antiparallel packed homodimers or heterodimers. Certain conditions, particularly the extracellular milieu, contribute to the formation of oligomers and multimers of some S100 proteins. For example, S100A8/S100A9 forms heterotetramers,6 7 S100A12 hexamers,6 8 and S100A4 forms tetramers and higher.9 Interaction of S100 proteins with several target proteins might explain their implication in a wide range of cellular events. Most S100 proteins are synthesised and localised intracellularly and participate in the regulation of a variety of processes within the cell, such as proliferation and differentiation, apoptosis, extracellular matrix remodelling and cell motility.5 Furthermore, several S100 proteins can be released from cells and their extracellular functions have been demonstrated. Once released into the extracellular space, they exert cytokine-like activities. Some S100 proteins can trigger intracellular signalling pathways through the receptor for advanced glycation end products (RAGE).10 11 For others, including S100A4, reliable receptor(s) remain to be identified.12
A number of human diseases are associated with an altered expression of S100 proteins. For example, certain S100 proteins have been studied in tumours (S100A1, S100A4, S100A6, S100A7, S100P and S100B), neurodegenerative diseases (S100B, S100A6, S100A12), cardiovascular diseases (S100A8/9, S100A4, S100A1), pulmonary diseases (S100A4), inflammatory diseases (S100A8/9, S100A12, S100A7, S100A4), diabetes mellitus (S100A12) and allergies (S100A12) (for a review see Donato).5 13-22 Several S100 proteins, including S100A2, S100A4, S100A6, S100A7, S100P and S100B, have been found to be involved in tumour progression, while others, such as S100A8, S100A9 and S100A12 are more likely related to inflammation. Since chronic inflammatory processes perpetuated by the immune system may abet tumour development,23 it could be suggested that certain members of the S100 protein family might represent a bridge between cancer and inflammation. Although the association between S100A4 and cancer promoting properties has been established,24-26 its potential role in chronic inflammatory diseases such as rheumatoid arthritis (RA) is a more recent discovery.9 27 28 Here, we summarise the current knowledge regarding S100A4 and its relation to the invasive behaviour of cells involved in the destructive process of chronic inflammation.
EXPRESSION AND FUNCTION OF S100A4
S100A4 (also known as metastasin, pEL, p9Ka, FspI, calcium protein placental homolog (CAPL) and calvasculin) is a small 11-kDa protein that was originally isolated as a gene differentially expressed in highly metastatic mouse mammary adenocarcinoma cells.29 Subsequently, it was also found in normal tissues,30 31 however the physiological function of S100A4 is not yet understood. In rats, intracellular S100A4 protein was detected in smooth muscle and endothelial cells of blood vessels, epithelial cells, brown adipose and liver tissue, parietal cells of the stomach, neuronal cells and some immune cells within the spleen, thymus and bone marrow.31 32 There is evidence that the expression of S100A4 is most likely related to different cancerous and immune processes.33 However, S100A4 has been found in normal human and rat tissues in fractions of T lymphocytes, and neutrophils, while its expression is weak in monocytes, hair follicles and some others. S100A4 protein has also been detected under chronic inflammatory conditions in macrophages, mast cells, neutrophils, certain T cells, dendritic cells and pericytes as well as activated synovial fibroblasts.9
Like other S100 proteins, S100A4 exerts intracellular and extracellular effects and is involved in a number of cellular events. Several studies have demonstrated that S100A4 may exist as homodimers (S100A4/S100A4) or heterodimers (S100A4/S100A1) and also has the potential to form oligomers.34 35 Since no enzymatic activity has been associated with S100A4, it is likely that interactions with other target proteins are critical for S100A4 activity. Intracellularly, S100A4 binds to several target molecules including the heavy chain of non-muscle myosin II36 and liprin β1,37 thereby modulating cell motility and adhesion. Furthermore, it interacts with the tumour suppressor protein p53 that may provide a link between S100A4 and apoptosis.30 36 38 39 Extracellular S100A4 has also been documented to stimulate neurite outgrowth of primary hippocampal neurons40 and the migration of astrocytic tumour cells.41 Moreover, S100A4 is involved in angiogenesis42 and remodelling of the extracellular matrix by means of upregulation of proteolytic enzymes.28 43 In this regard Duarte et al44 proposed that S100A4 is a novel negative regulator of matrix mineralisation that modulates the process of osteoblast differentiation.
PROPOSED FUNCTION OF S100A4 ASSOCIATED WITH TUMOUR PROGRESSION AND METASTASIS
In a complex cascade of events in tumourigenesis and metastatic progression, a variety of regulatory molecules are involved at different stages. Involvement of S100A4, as one of the regulatory elements, has been demonstrated in transgenic mice26 as well as in humans.24 25 Both intracellular (interaction with target proteins) and extracellular (cytokine-like triggering of signal transduction) forms of S100A4 contribute to the metastasis-promoting function of the protein.
Intracellular S100A4 is known to interact with cytoskeletal components including the heavy chain of non-muscle myosin, non-muscle tropomyosin and F-actin,30 36 39 thereby affecting the motility of cancer cells. Interaction of S100A4 with the heavy chain of non-muscle myosin (MHC) results in inhibition of protein kinase C (PKC) and casein kinase (CK)-2 dependent phosphorylation of MHC.45 46 This interaction increases the solubility of myosin and regulates cytoskeletal dynamics. Similarly, binding of S100A4 to non-muscle tropomyosin is also thought to be responsible for the disassembly of actin filaments.30 These data suggest that S100A4 can modulate the invasiveness of tumour cells. While these interactions have been documented to be calcium-dependent, there are some reports on calcium-independent interactions of S100A4 (for review see Santamaria-Kisiel et al).3
Besides cell motility, cell–cell adhesions are considered to be another property of metastatic cells. Cooperation between S100A4 and E-cadherin (transmembrane glycoprotein that mediates Ca2+ dependent cell–cell adhesion) has been studied in mouse tumour cells as well as in humans. In both cases, an inverse correlation in the expression of E-cadherin and S100A4 was demonstrated, suggesting that the invasiveness of tumours expressing S100A4 could be induced by the abrogation of E-cadherin expression.47 48 Moreover, it has been proven that S100A4 contributes to cell adhesion via binding to liprin β1 (transmembrane tyrosine phosphatase-interacting protein), thus modulating LAR (transmembrane phosphotyrosine phosphatase)-dependent signalling directly involved in cell adhesion.37
S100A4 has also been reported to regulate proliferation and apoptosis. Recently, it was shown that the expression of S100A4 is associated with an increased amount of p53, however the conformational form of p53 was not studied.49 Moreover, physical interaction of S100A4 with the C-terminal regulatory domain of tumour suppressor protein p53 has been shown previously.38 Induction of S100A4 in cell lines expressing wild-type p53 modulated the expression of p53 downstream target genes including p21/WAF and bax.38 It has been suggested that S100A4 can enhance p53-dependent apoptosis and thereby accelerate the loss of wild-type p53 functions, and consequently contribute to the development of a more aggressive phenotype during early tumour progression. Moreover, a reduced frequency of apoptosis was observed in the spleen of S100A4–/– animals after whole-body γ irradiation compared to wild-type animals.50 By contrast, extracellular S100A4 has been found to downregulate bax in mouse adenocarcinoma cells, which might increase tumour cell survival. Oligomeric and dimeric forms of S100A4 exerted equal inhibitory effect on the transcription of bax, however, no influence on the expression of p21/waf was observed.51
Another important hallmark of invasive tumour growth and cancer metastasis is angiogenesis. S100A4 has been found to exert its role in angiogenesis particularly via the modulation of the expression of thrombospondin 1 and matrix metalloproteinases (MMPs). Thrombospondin 1 belongs to the class of extracellular matrix glycoproteins with antiangiogenic effects that can repress the progression of tumours.52 Thus, the treatment of tumour and endothelial cells with S100A4 oligomer-induced downregulation of thrombospondin 1 gene expression.38 51 Moreover, the S100A4 oligomer was capable of stimulating angiogenesis by promoting the chemotactic motility of endothelial cells in vitro, and of inducing corneal neovascularisation in vivo.51 Dysregulation of MMPs participates in remodelling of the extracellular matrix, tumour cell migration and invasion.53 An association between S100A4 and the production of MMPs has also been reported.38 43 51 Extracellular S100A4 strongly stimulated the proteolytic activity in endothelial and tumour cells, particularly by upregulating MMP13. In a highly metastatic osteosarcoma cell line transfected with a specific ribozyme against the S100A4 gene transcript, the downregulation of S100A4 expression resulted in a reduction of the mRNA levels of MMP2, MMP14 and the endogenous tissue inhibitor TIMP-1.54 In addition, the suppression of S100A4 has been shown to significantly reduce the expression and proteolytic activity of MMP9.55
Summarising the above-mentioned data, one can support the idea that the metastasis-inducing S100A4 plays a pivotal role in modulating the tumour stroma51 because it has been shown that S100A4 is involved in the regulation of cancer invasiveness and metastasis (summarised in Helfman et al).56 Intracellular as well as extracellular functions of S100A4 have been proposed to be involved in this process, and the protein has been suggested as a prognostic marker for several tumours.33
PROPOSED FUNCTION OF S100A4 IN RHEUMATOID ARTHRITIS
Rheumatoid arthritis (RA) is an autoimmune disease characterised by bone and cartilage destruction, chronic synovial inflammation and hyperplasia. In recent years, synovial fibroblasts (SF) have been suggested to play an active role in the pathogenesis of RA.57 58 Masuda et al59 found certain genes, including S100A4, to be upregulated in proliferating compared with non-proliferating RA SFs. Moreover, S100A4 mRNA has been detected in the lining as well as sublining layer of RA synovial tissues, while its expression was not observed in healthy synovial tissues. Recently, we have demonstrated a strong upregulation of S100A4 protein in RA compared with osteoarthritis (OA) and control synovial tissues.9 28 Most importantly, the expression of S100A4 protein has been detected at sites of cartilage and bone destruction. Synovial fibroblasts, as well as several immune and vascular cells, were identified to produce S100A4.9 In comparison with other S100 proteins such as S100A7, S100A8, S100A9 and S100A12, the majority of cells in RA synovial tissue have been found to express S100A4. The upregulation of S100A4 in RA synovial tissue was consistent with the finding of rather high concentrations of the protein in RA compared with OA plasma (1100 vs 211 ng/ml) and synovial fluid (1980 vs 247 ng/ml) (fig 1). In the plasma and synovial fluid of patients with RA, S100A4 exists in a bioactive oligomeric conformation whereas in OA the majority of extracellular S100A4 was detected in the less active, dimeric form. Consistent with observations in tumour models, extracellular S100A4 induced the upregulation of several MMPs such as MMP1, MMP3, MMP9 and MMP13,28 and stabilised the tumour suppressor protein p53 in RA SFs.9 The active oligomeric S100A4 protein also revealed notable effects on the transcriptional regulation of p53 target genes including Bcl-2, p21/WAF and HDM2 that are involved in proliferation and apoptosis.
In summary, S100A4 is increased in patients with RA locally at sites of inflammation, as well as systemically in circulation. Moreover, expression of S100A4 is localised specifically at sites of joint destruction and was shown to modulate in vitro proliferation, apoptosis and the production of several MMPs in RA synovial fibroblasts (fig 2).
Recently, the potential involvement of S100A4 in the pathogenesis of OA has also been studied.60 Enhanced expression of S100A4 compared with normal tissue was detected by immunohistochemistry and Western blotting in cartilage from patients with OA and in a human articular chondrocyte cell line. Yammani et al demonstrated that S100A4 binds to multiligand RAGE and thereby stimulates the expression of MMP13 in human articular chondrocytes. Activation of RAGE by S100A4 in chondrocytes triggers signal transduction pathways stimulating phosphorylation of Pyk-2 and mitogen activated protein (MAP) kinases (ERK-1/2, p38 and JNK), the activation of nuclear factor (NF)κB and the production of reactive oxygen species (ROS). These data suggest that S100A4 RAGE signalling might be involved in the process of cartilage degeneration in OA.60 It appears that S100A4 is involved in numerous processes leading to joint destruction.
PROPOSED FUNCTION(S) OF OTHER S100 PROTEINS IN RHEUMATOID ARTHRITIS
Within the S100 protein family a subgroup of phagocyte-specific proteins (calgranulins) has been identified. The three members of this subgroup, S100A8, S100A9 and S100A12, are predominantly expressed in cells of myeloid origin and exert proinflammatory functions in the extracellular milieu via interaction with RAGE.11 61-63 Their elevated levels have been associated with a number of inflammatory diseases, particularly with RA.
S100A12 (calgranulin C or EN-RAGE) is mainly detected in granulocytes64 and in lower levels in monocytes.3 However, increased levels of S100A12 in inflamed tissues have also been documented in a number of disorders including psoriasis,16 65 inflammatory bowel diseases,11 66 Kawasaki disease,67 68 giant cell arteritis,69 type 2 diabetes17 and Alzheimer disease.70
The results from in vitro studies and animal models have revealed the impact of S100A12 in the pathogenesis of chronic arthritis. S100A12 has been immunodetected within the synovial sublining layer, the perivascular region and the synovial lining layer in RA. Staining of sequential sections of the synovial lining layer by CD68 confirmed occasional S100A12 positive macrophages.71 S100A12 may stimulate the accumulation of neutrophils by inducing their release from the bone marrow as well as by activating their adhesion and migration toward inflammatory sites.72 In the early stages of inflammation, binding of S100A12 to RAGE may lead to stimulation of endothelial cells by increasing surface expression of adhesion molecules (vascular cell adhesion molecule (VCAM)-1 and intercellular adhesion molecule (ICAM-1)) and to promotion of transendothelial migration of phagocytes.73 Furthermore, S100A12 upregulates NFκB driven transcription of some inflammatory cytokines such as tumour necrosis factor (TNF)-α in inflammatory cells.11 Increased levels of S100A12 were found in synovial fluid and plasma from patients with RA, gout and psoriatic arthritis, however, they were undetectable in patients with non-inflammatory disorders such as OA.72 Foell et al74 found in juvenile idiopathic arthritis (JIA) that S100A12 serum concentration correlates with disease activity, and decreases in response to different anti-inflammatory therapies. Moreover, elevated serum S100A12 could be detected weeks before clinically apparent relapses in patients with previously well controlled disease.
The other calgranulins, S100A8 (migration inhibitory factor-related protein (MRP)8 or calgranulin A) and S100A9 (MRP14 or calgranulin B) exist in general as a heterodimer and are actively secreted by human monocytes and activated granulocytes.75 The expression of S100A8 and S100A9 without concomitant formation of their complex has been rarely found, eg, in a chronic type of inflammatory reaction in glomerulonephritis or in chronic renal allograft rejection.76 77 Upregulation of the S100A8/9 complex, known as calprotectin, however, has been well described in inflammatory disorders such as cystic fibrosis,78-81 psoriasis,82 tuberculosis,83 Crohn disease and ulcerative colitis.66 The presence of S100A8 and S100A9 in infiltrated macrophages in RA was first detected in 1987.84 The authors demonstrated that resting normal tissue macrophages did not express MRP8 and MRP14, macrophages in acutely inflamed tissues expressed MRP14 but not MRP8, and in chronic inflammation, infiltrated macrophages expressed both proteins. Additionally, Youssef et al found predominant expression of MRP8, MRP14 and their heterodimer at the site of cartilage destruction in RA.85 Analysis of blood and synovial fluid showed significantly elevated levels of S100A8/9 in plasma and synovial fluid of patients with RA compared with OA.72 86 Moreover, concentrations of the dimer were significantly higher in synovial fluid than in serum, and they correlated with each other showing a strong association with disease activity; it was therefore suggested that MRP8/14 could be parameters of prognostic value for further disease flare in patients with JIA.87 88 In vitro, the S100A8/A9 heterodimer may enhance inflammation through increased production of proinflammatory cytokines including TNFα.89 In addition to chronic inflammation, most recent data point to the fact that the MRP8/14 dimer is associated with the development of acute coronary syndrome. Because levels of the dimer increase in blood before the levels of markers of myocardial necrosis, the MRP8/14 dimer might serve as a novel sensitive predictor of the syndrome.14 Based on the data above, one can suggest that S100A12 and S100A8/A9 proteins are valuable markers of inflammation as well as potential targets for future therapies.
CONCLUSIONS
Over the past decade, many reports have produced great strides forward in understanding the role of S100A4 in tumour cell invasion and subsequent metastasis. As shown by several research groups, the interaction of the protein with its several effectors can activate signalling pathways and modulate detachment of the extracellular matrix, adhesion, cell motility, angiogenesis, cell proliferation and apoptosis. Interestingly, hallmarks of metastatic activity of tumour cells and the invasive behaviour of synovial fibroblasts in RA appear similar. Analogous to the environment of tumours, levels of S100A4 have been been found to be increased in inflamed synovial tissue and body fluids in patients with RA. S100A4 is proposed to be upregulated upon synovial fibroblast activation/proliferation59 as well as by several inflammatory cells that accumulate in RA synovial tissue.9 28 Therefore, we suggest that the S100A4 properties are not clearly RA-specific, however, can be related to systemic inflammation or/and activation of synovial fibroblasts. Based on in vitro studies, S100A4 can upregulate the production of matrix degrading enzymes and stabilise the tumour suppressor protein p53 in synovial fibroblasts.9 28 Thus, it can be suggested that S100A4 takes part in stimulating cell proliferation, modulating apoptosis and tissue remodelling, thereby contributing to the process of inflammation and tissue destruction. Further analysis and characterisation of the molecular mechanisms of S100A4 regulating pathways are needed to clarify the contribution of S100A4 to the immune/inflammatory processes leading to synovial hyperplasia and inflammation.
REFERENCES
Footnotes
Funding: This study was supported by the IGA MZ CR, grant no: NR/9082-4.
Competing interests: None declared.
Ethics approval: Ethics approval was obtained.