Friend or Foe: S100 Proteins in Cancer

S100 proteins are widely expressed small molecular EF-hand calcium-binding proteins of vertebrates, which are involved in numerous cellular processes, such as Ca2+ homeostasis, proliferation, apoptosis, differentiation, and inflammation. Although the complex network of S100 signalling is by far not fully deciphered, several S100 family members could be linked to a variety of diseases, such as inflammatory disorders, neurological diseases, and also cancer. The research of the past decades revealed that S100 proteins play a crucial role in the development and progression of many cancer types, such as breast cancer, lung cancer, and melanoma. Hence, S100 family members have also been shown to be promising diagnostic markers and possible novel targets for therapy. However, the current knowledge of S100 proteins is limited and more attention to this unique group of proteins is needed. Therefore, this review article summarises S100 proteins and their relation in different cancer types, while also providing an overview of novel therapeutic strategies for targeting S100 proteins for cancer treatment.


The S100 Family
In 1965, Blake W. Moore isolated a protein from the bovine brain to identify specific proteins of the nervous system. Since it was soluble in 100% saturated ammonium sulfate at neutral pH, the protein was named "S-100" [1]. Isobe et al. later showed that this S100 protein formed a dimer, consisting of two homologous yet different molecules, defined as S100A and S100B [2,3]. Further research from the past decades revealed that their expression was not limited to nervous tissue but that S100 proteins are found in various tissues exclusively in vertebrates. Subsequently, many more S100-related proteins were discovered, and currently, 25 family members are known [4]. It has since been shown that these proteins are involved in a variety of different pathways, thereby playing a critical role in essential cellular processes, such as proliferation, apoptosis, differentiation, and inflammation [5].

S100 Proteins in Cancer
In the 1980s, soon after the identification of the first S100 family members, the presence of S100 proteins was also observed in different cancer types, drawing rising attention to this group of calcium-binding proteins within the oncological field. In this context, S100 expression was, for example, detected in metastatic melanoma [70,71], renal cell carcinoma [72], and breast cancer [73]. Moreover, an S100 protein expressed in mouse Ehrlich ascites-tumour cells was identified as S100A6 [74], and S100P was found in the cell nuclei of brain tumours [75]. Further research from the past decades Cancers 2020, 12, 2037 5 of 36 demonstrated that dysregulation of S100 proteins, typically upregulation, can be connected to tumour growth, angiogenesis, and metastasis [76]. Several S100 proteins could also be linked to drug resistance and seem to be involved in mediating the response to chemotherapy [77]. Each cancer type shows a specific S100 expression profile, and the different S100 family members function individually in a tissue-dependent manner. S100A2, for example, operates as a tumour suppressor in oral cancer while it promotes tumour growth in lung cancer [78]. The role of S100 proteins has been widely examined in breast cancer, lung cancer, and melanoma, but pathologic S100 signalling could also be observed in additional cancer types, such as ovarian, colorectal, and pancreatic cancer.
Zhang et al. analysed S100 mRNA expression in breast cancer patients (n = 1117), using the online databank Gene Expression Omnibus. Intriguingly, the expression of S100 proteins correlates with the overall survival (OS) of patients, depending on cancer subtype and clinicopathologic features. For example, high mRNA levels of S100A8 and S100A9 predict worse prognosis in the luminal A-type breast cancer, while increased levels of S100A10, S100P, and S100Z showed shorter OS in patients with the basal-like subtype. In contrast, longer OS was observed for patients with luminal a subtype expressing high levels of S100A1, S100A2, and S100A6 and for patients with the basal-like subtype expressing high levels of S100A14. Overall, S100A1 and S100A6 seem to predict better prognosis and S100A8, S100A9, S100A11, and S100P seem to worsen OS rates [84]. Zhong et al. found a correlation between S100A8 and lymph node metastasis in breast cancer and therefore, tumour progression [85]. Further publications covering this connection identified S100B as a serum marker in endocrine-resistant breast cancer [86] and elevated S100B serum levels as a negative prognostic value for breast cancer [87,88]. Nevertheless, a collective expression pattern of all S100 members together might have higher prognostic value than the single proteins [82]. S100A4 is highly involved in metastasis and invasion of many different cancer types, including breast cancer [89,90]. Barraclough et al. early showed that S100A4 is associated with a metastatic phenotype using a rat mammary model. It was suggested that the underlying mechanism includes the interaction of S100A4 with cytoskeletal proteins, such as actin and myosin, thereby promoting cell motility [91]. These findings are supported by further studies, showing that intracellular S100A4 interacts with actin, non-muscle myosin heavy chain IIA (NMIIA), and tropomyosin, leading to cell migration [92,93]. Although S100A1 does not seem to be directly involved in breast cancer signalling, it was shown that S100A1 reduces the activity of intracellular S100A4 [94]. The interaction between S100A4 and matrix metalloproteinase 2 (MMP2) induces epithelial-mesenchymal transition (EMT), which is considered to be an initial step during invasion and metastasis [95]. S100A4 can not only be found in cancer cells but is also highly expressed in stromal cells of the tumour microenvironment (TME), such as fibroblasts, T-cells, macrophages, and neutrophils [81,96]. Extracellular S100A4 in the TME induces the release of pro-inflammatory factors (e.g., interleukin 6 (IL-6), interleukin 8 (IL-8), and C-X-C motif chemokine 10 (CXCL10)), which then converts monocytes into tumour-associated macrophages (TAMs), resulting in metastasis and drug resistance [97].
Only a few studies have been performed to investigate the signalling of S100A6 in breast cancer. However, S100A6 induces the activity of calcyclin-binding protein/Siah-1-interacting protein (Cacy/SIP), which is involved in tumour invasion and metastasis in breast cancer most likely by increasing β-catenin levels [98,99]. Another exciting approach showed that S100A6 levels decrease in human breast cancer cell line MDA-MB-231 when treated with anti-tumour agents, namely tumour Cancers 2020, 12, 2037 6 of 36 necrosis factor-related apoptosis-inducing ligand (TRAIL) and etoposide, indicating that S100A6 might be a potential biomarker for apoptosis during therapy [100]. S100A7 was also identified as an essential protein in breast cancer cells, especially in estrogen receptor α (ERα) negative cells, where the binding to RAGE leads to activation of extracellular signal-regulated kinase (ERK) and NF-κB signalling, resulting in tumour growth and enhanced metastasis [101]. Moreover, increased NF-κB activity was observed in S100A7-overexpressing breast cancer cells, associated with evaluated levels of NF-κB target genes matrix metalloproteinase 9 (MMP9) and vascular endothelial growth factor (VEGF), resulting in proliferation and invasion [102]. Further, the binding of S100A7 and RAGE leads to the recruitment of TAMs, which then promote further tumour growth, angiogenesis, and metastasis by expressing chemokine (C-C motif) ligand 2 (CCL2), cyclooxygenase-2 (COX2), and VEGF [101,103]. Intracellular S100A7 interacts with the transcriptional cofactor constitutive photomorphogenesis 9 (COP9) constitutive photomorphogenic homolog subunit 5 (COPS5), also known as c-Jun activation domain-binding protein-1 (Jab1), which then accumulates in the nucleus and promotes the expression of AP-1 and NF-κB, resulting in enhanced tumour growth and invasion [104].
The binding of S100A8/S100A9 to RAGE promotes breast cancer cell growth by inducing MAPK signalling. In addition, RAGE then mediates cell migration by promoting actin polymerisation and EMT, resulting in metastasis and invasion [105]. Although not all S100 proteins have been studied intensively in this context, the current knowledge of S100 signalling in breast cancer is summarised in Figure 1. Only a few studies have been performed to investigate the signalling of S100A6 in breast cancer. However, S100A6 induces the activity of calcyclin-binding protein/Siah-1-interacting protein (Cacy/SIP), which is involved in tumour invasion and metastasis in breast cancer most likely by increasing β-catenin levels [98,99]. Another exciting approach showed that S100A6 levels decrease in human breast cancer cell line MDA-MB-231 when treated with anti-tumour agents, namely tumour necrosis factor-related apoptosis-inducing ligand (TRAIL) and etoposide, indicating that S100A6 might be a potential biomarker for apoptosis during therapy [100]. S100A7 was also identified as an essential protein in breast cancer cells, especially in estrogen receptor α (ERα) negative cells, where the binding to RAGE leads to activation of extracellular signalregulated kinase (ERK) and NF-κB signalling, resulting in tumour growth and enhanced metastasis [101]. Moreover, increased NF-κB activity was observed in S100A7-overexpressing breast cancer cells, associated with evaluated levels of NF-κB target genes matrix metalloproteinase 9 (MMP9) and vascular endothelial growth factor (VEGF), resulting in proliferation and invasion [102]. Further, the binding of S100A7 and RAGE leads to the recruitment of TAMs, which then promote further tumour growth, angiogenesis, and metastasis by expressing chemokine (C-C motif) ligand 2 (CCL2), cyclooxygenase-2 (COX2), and VEGF [101,103]. Intracellular S100A7 interacts with the transcriptional cofactor constitutive photomorphogenesis 9 (COP9) constitutive photomorphogenic homolog subunit 5 (COPS5), also known as c-Jun activation domain-binding protein-1 (Jab1), which then accumulates in the nucleus and promotes the expression of AP-1 and NF-κB, resulting in enhanced tumour growth and invasion [104].
The binding of S100A8/S100A9 to RAGE promotes breast cancer cell growth by inducing MAPK signalling. In addition, RAGE then mediates cell migration by promoting actin polymerisation and EMT, resulting in metastasis and invasion [105]. Although not all S100 proteins have been studied intensively in this context, the current knowledge of S100 signalling in breast cancer is summarised in Figure 1. . S100 signalling in breast cancer. Intracellular S100A4 interacts with cytoskeletal proteins, such as actin, non-muscle myosin heavy chain IIA (NMIIA), and tropomyosin, which promotes cell motility. Besides, S100A4 can induce epithelial-mesenchymal transition (EMT) by regulating the expression of matrix metallopeptidases 2 (MMP2), leading to invasion and metastasis. Extracellular S100A4, located in the tumour microenvironment (TME), induces the release of pro-inflammatory factors (e.g., IL-6, IL-8, and CXCL10). These cytokines then convert monocytes into tumour-associated macrophages (TAMs), which in return, promote EMT, proliferation, and drug resistance of the tumour cells. The binding of extracellular S100A7 to the receptor for advanced glycation end products (RAGE) induces mitogen-activated protein kinase (MAPK) and nuclear factor "kappa-light-chain- Figure 1. S100 signalling in breast cancer. Intracellular S100A4 interacts with cytoskeletal proteins, such as actin, non-muscle myosin heavy chain IIA (NMIIA), and tropomyosin, which promotes cell motility. Besides, S100A4 can induce epithelial-mesenchymal transition (EMT) by regulating the expression of matrix metallopeptidases 2 (MMP2), leading to invasion and metastasis. Extracellular S100A4, located in the tumour microenvironment (TME), induces the release of pro-inflammatory factors (e.g., IL-6, IL-8, and CXCL10). These cytokines then convert monocytes into tumour-associated macrophages (TAMs), which in return, promote EMT, proliferation, and drug resistance of the tumour cells. The binding of extracellular S100A7 to the receptor for advanced glycation end products (RAGE) induces mitogen-activated protein kinase (MAPK) and nuclear factor "kappa-light-chain-enhancer" of activated B-cells (NF-κB) signalling, resulting in tumour growth and metastasis. Increased NF-κB activity was observed in S100A7-overexpressing breast cancer cells, associated with evaluated levels of matrix metalloproteinase 9 (MMP9) and vascular endothelial growth factor (VEGF), resulting in proliferation and invasion. The binding of S100A7 and RAGE also leads to the recruitment of TAMs, which then promote further tumour growth, angiogenesis, and metastasis by expressing chemokine (C-C motif) ligand 2 (CCL2), cyclooxygenase-2 (COX2), and VEGF. Intracellular S100A7 interacts with the transcriptional cofactor COP9 constitutive photomorphogenic homolog subunit 5 (COPS5), which in turn promotes the expression of AP-1 and NF-κB, resulting in enhanced tumour growth and invasion. S100A8/S100A9 enhances breast cancer cell growth by inducing MAPK signalling in a RAGE-dependent manner. In addition, RAGE mediates cell migration by promoting actin polymerisation and EMT, leading to metastasis and invasion.

S100 Proteins in Lung Cancer
Lung cancer is the most common cancer type with a high mortality rate [79], mostly due to lacking diagnostic means for efficient early detection [106]. There are two major groups of lung cancer: non-small cell lung cancer (NSCLC), which is diagnosed in about 80% of all lung cancer patients, and small cell lung cancer (SCLC) [107,108]. It was shown that S100A1, S100A2, S100A3, S100A4, S100A6, S100A7, S100A8/S100A9, S100A10, S100A11, S100B, and S100G are overexpressed in NSCLC, and except S100B, all are predictive for poor survival [109]. The function of several S100 proteins in lung cancer remains unclear. However, some family members have been studied in this context and are summarised in Figure 2.
The studies on S100A2 in NSCLC presented conflicting data [110]. For example, Feng et al. showed that S100A2 was downregulated in NSCLC cell lines and detected CpG methylation in the promoter region of the S100A2 gene, indicating that S100A2 was suppressed during early-stage carcinogenesis [111]. In contrast, Heighway et al. demonstrated that S100A2 was strongly expressed in primary NSCLC tissue [112], and Bulk et al. later showed that S100A2 acts as a metastasis inducer in mouse models [113]. In 2014, Hountis et al. introduced a dual role concept for S100A2 in lung cancer, implicating that S100A2 is primarily expressed in the nucleus at an early stage of NSCLC, where it mediates resistance to p53-dependent apoptosis and inhibits tumour-promoting genes (e.g., PA1-1 and vimentin) and in later stages relocates to the cytoplasm in a Ca 2+ -dependent manner [114]. S100A4 seems to be an important player in the development and metastasis of lung cancer, promoting tumour cell proliferation and motility [115,116]. The overexpression of S100A4 in lung adenocarcinoma (ADC), a subtype of NSCLC, was linked to reduced OS of these patients [117]. In vitro studies revealed that extracellular S100A4 inhibited autophagy and induced Wnt signalling in a RAGE-dependent manner and intracellular S100A4 additionally activated β-catenin, resulting in increased proliferation and enhanced viability of lung cancer cells [118]. Further, in vitro studies showed that the depletion of S100A4 resulted in decreased NF-κB signalling and inhibition of matrix metalloproteinases 9 (MMP9), demonstrating that the S100A4/NF-κB/MMP9 signalling axis promotes invasion [117].
Regarding the role of S100A7 in lung cancer, it was shown that this protein is highly expressed in lung squamous cell carcinoma (SCC), an NSCLC subtype, and that knockdown of S100A7 resulted in decreased NF-κB-dependent cell proliferation [119]. In vitro, Wang et al. demonstrated that S100A7 expression was induced by the Hippo pathway and depletion of S100A7 lead to suppression of DNp63, a marker for SSC, while the ADC markers TFF1 and napsin A were upregulated, and an inverse correlation of S100A7 and yes-associated protein (YAP) was observed. In summary, these data implicate that S100A7 is an essential player in cellular plasticity and ADC to SSC transdifferentiation [120]. S100B was introduced as a possible biomarker for brain metastasis in lung cancer patients [121,122]. Further investigation of S100B in an ADC-derived brain metastasis cell line showed that depletion of S100B resulted in the downregulation of pro-survival factors, namely B-cell lymphoma 2 (Bcl-2) and B-cell lymphoma extra-large (Bcl-xL), both inhibitors of apoptosis, indicating that S100B suppresses apoptosis in these cancer cells. In this study, it was moreover observed that S100B correlates with enhanced proliferation, migration, and invasion [123]. In contrast, Liu et al. showed that increased mRNA levels of S100B predicted better OS in patients with NSCLC [109]. Therefore, the molecular Lung cancer is the most common cancer type with a high mortality rate [79], mostly due to lacking diagnostic means for efficient early detection [106]. There are two major groups of lung cancer: non-small cell lung cancer (NSCLC), which is diagnosed in about 80% of all lung cancer patients, and small cell lung cancer (SCLC) [107,108]. It was shown that S100A1, S100A2, S100A3, S100A4, S100A6, S100A7, S100A8/S100A9, S100A10, S100A11, S100B, and S100G are overexpressed in NSCLC, and except S100B, all are predictive for poor survival [109]. The function of several S100 proteins in lung cancer remains unclear. However, some family members have been studied in this context and are summarised in Figure 2. Figure 2. S100 signalling in lung cancer. Extracellular S100A4 inhibits autophagy and induces Wnt signalling by interacting with the receptor for advanced glycation end products (RAGE) and intracellular S100A4 additionally activates β-catenin, resulting in increased proliferation and enhanced viability of lung cancer cells. S100A4 also induces the expression of (MMP9) by activating nuclear factor "kappa-light-chain-enhancer" of activated B-cells (NF-κB), thereby promoting invasion and metastasis. S100A7 is most likely involved in adenocarcinoma (ADC) to squamous cell carcinoma (SSC) transdifferentiation of lung cancer cells, by upregulating the SSC marker DNp63 and Figure 2. S100 signalling in lung cancer. Extracellular S100A4 inhibits autophagy and induces Wnt signalling by interacting with the receptor for advanced glycation end products (RAGE) and intracellular S100A4 additionally activates β-catenin, resulting in increased proliferation and enhanced viability of lung cancer cells. S100A4 also induces the expression of (MMP9) by activating nuclear factor "kappa-light-chain-enhancer" of activated B-cells (NF-κB), thereby promoting invasion and metastasis. S100A7 is most likely involved in adenocarcinoma (ADC) to squamous cell carcinoma (SSC) transdifferentiation of lung cancer cells, by upregulating the SSC marker DNp63 and downregulation of the ADC markers thyroid transcription factor 1 (TTF1) and aspartic proteinase napsin (napsin A). In this context, an inverse correlation of S100A7 and yes-associated protein (YAP) was observed. Moreover, S100A7 seems to activate NF-κB-dependent cell proliferation. Within lung cancer-derived brain metastasis cells, S100B was shown to upregulate the expression of B-cell lymphoma 2 (Bcl-2) and B-cell lymphoma extra-large (Bcl-xL), indicating that S100B is capable of suppressing apoptosis.
As previously discussed, S100A4 is involved in the metastatic processes in other cancer types, and similar observations were made in melanoma. Upregulation of S100A4 correlates with the upregulation of RAGE in metastatic melanoma cells in vitro, and elevated levels of S100A4 and RAGE are associated with high tumour burden in vivo [127,128]. It was shown that extracellular S100A4 activates NF-κB in a RAGE-dependent manner, resulting in the release of TNFα [129]. S100A4 also induces the secretion of further paracrine factors, such as IL-8 and CCL2, which in return, promote angiogenesis and recruitment Cancers 2020, 12, 2037 9 of 36 of monocytes, creating an inflammatory milieu in the tumour microenvironment [130]. Moreover, it was revealed that extracellular S100A4 decreases the expression of occludin and VE-cadherin in endothelial cells (ECs), thereby disrupting cell-cell adhesion. This enables melanoma cells to transmigrate through the EC monolayer into the bloodstream [128]. In summary, S100A4 induces metastatic signalling and promotes cell migration in malignant melanoma.
When monitoring stage IV melanoma patients during immunotherapy with ipilimumab, the heterodimer S100A8/S100A9 attracted attention, as high serum levels of it in early stages of treatment predicted worse response [131]. The release of S100A8/S100A9 can be induced by UV radiationexposed keratinocytes, and extracellular S100A8/S100A9 then promotes proliferation and migration of melanocytes via RAGE-dependent signalling [132]. It could be demonstrated that interaction between S100A8/S100A9 and RAGE leads to increased levels of the metalloproteinases MMP2, MMP9, and MMP14 in melanoma cells, which enhances metastatic properties [133]. Besides RAGE, novel ligands for S100A8/S100A9 were found. S100A8/S100A9 binds to the melanoma cell adhesion molecule (MCAM), a highly expressed cell adhesion molecule in melanoma [134]. This interaction activates mitogen-activated protein kinase kinase kinase 8 (MAP3K8), also termed tumour progression locus 2 (TPL2), which conversely stimulates the transcription factor ETS translocation variant 4 (ETV4), leading to induction of MMP25 and promoting melanoma lung metastasis [135]. The homodimer of S100A9 additionally interacts with extracellular matrix metalloprotease inducer (EMMPRIN), which activates TNF receptor-associated factor (TRAF2)-dependent NF-κB signalling and the upregulation of cytokines such as TNFα, CXCL1, CXCL2, and CXCL3, resulting in metastasis [136]. MCAM and EMMPRIN are both highly expressed in melanoma and inhibition of the interaction between these receptors and S100A8/A9 could suppress lung metastasis in vivo, thereby representing an interesting target for therapeutic approaches [137]. S100B is described to be a prognostic marker for the advanced stages of melanoma, especially for distant metastasis, and increased serum S100B levels are associated with shorter disease-free survival (DFS) and OS [138][139][140]. Investigating the molecular mechanisms of S100B in melanoma cells, it could be revealed that S100B interacts with ribosomal S6 kinase (RSK). RSK is a downstream target of ERK, and S100B inhibits phosphorylation of RSK by ERK in a Ca 2+ -dependent manner so that RSK remains in the cytoplasm and does not relocate to the nucleus. This results in increased activity of RSK in the cytoplasm, leading to improved tumour survival [141]. In vitro studies also revealed that increased levels of p53 in malignant melanoma cells promote expression of S100B, and within a negative feedback loop, S100B inhibits p53 activity at the protein level, thereby preventing p53-dependent apoptosis [54,142]. While p53 is mutated in most cancer types, 80% of melanomas express wild-type p53 [143], and therefore S100B might be an attractive target for new therapeutic approaches in this context. An overview of S100 signalling in malignant melanoma is given in Figure 3. . S100 signalling in melanoma. Extracellular S100A4 binds to the receptor for advanced glycation end products (RAGE) and thereby activates nuclear factor "kappa-light-chain-enhancer" of activated B-cells (NF-κB), resulting in the release of cytokines such as tumour necrosis factor α (TNFα). These cytokines then promote angiogenesis and recruitment of monocytes, creating an inflammatory milieu in the tumour environment. Furthermore, extracellular S100A4 also decreases the expression of occluding and vascular endothelial cadherin (VE)-cadherin in endothelial cells (ECs), thereby disrupting cell-cell adhesion and enabling the tumour cells to transmigrate through the EC monolayer into the bloodstream. The release of S100A8/S100A9 can be induced by UV radiation-exposed keratinocytes, and extracellular S100A8/S100A9 then promotes proliferation and migration of melanocytes via RAGE signalling. The interaction between S100A8/S100A9 and RAGE can lead to increased levels of the metalloproteinases MMP2, MMP9, and MMP14 in melanoma cells, thereby enhancing metastatic properties. In addition to RAGE, S100A8/A9 also binds to the melanoma cell adhesion molecule (MCAM), thereby activating tumour progression locus 2 (TPL2) and stimulating the transcription factor ETS translocation variant 4 (ETV4), leading to the induction of MMP25 and promoting melanoma lung metastasis. The homodimer of S100A9 additionally interacts with extracellular matrix metalloprotease inducer (EMMPRIN), which activates TNF receptorassociated factor (TRAF2)-dependent NF-κB signalling and the upregulation of cytokines such as TNFα, chemokine (C-X-C motif) ligand 1 (CXCL1), CXCL2, and CXCL3, resulting in metastasis. S100B inhibits the phosphorylation of ribosomal S6 kinase (RSK) by extracellular signal-regulated kinase (ERK) in a Ca 2+ -dependent manner so that RSK remains in the cytoplasm, leading to improved tumour survival. S100B also inhibits p53 activity at the protein level, thereby preventing p53-dependent apoptosis.

S100 Proteins in Ovarian Cancer
Recent studies investigating mRNA expression of S100 proteins in correlation with the survival of ovarian cancer patients demonstrated that high levels of the S100 family members S100A2, S100A10, S100A11, S100A15, S100A16, and S100B predict worse OS, while S100A1, S100A3, S100A5, S100A6, S100A13, S100G, and S100Z are associated with longer OS, partially depending on cancer subtype and clinicopathological features [144,145]. However, the underlying mechanisms of these S100 proteins in the context of ovarian cancer have not yet been fully understood.
Nevertheless, among others, S100A3, S100A10, and S100B were identified to be related to drug resistance in ovarian cancer [146][147][148]. Elevated expression of S100A3 was found primarily in ovarian cancer cells resistant to cisplatin and paclitaxel, both standard chemotherapeutic drugs, as well as topotecan, a chemotherapeutic agent commonly used in second-line treatment [146]. S100A10 was 1 of 11 genes associated with multidrug resistance in ovarian primary serous carcinoma tissue, although further studies are needed to validate these data, as this relation could not be confirmed in Figure 3. S100 signalling in melanoma. Extracellular S100A4 binds to the receptor for advanced glycation end products (RAGE) and thereby activates nuclear factor "kappa-light-chain-enhancer" of activated B-cells (NF-κB), resulting in the release of cytokines such as tumour necrosis factor α (TNFα). These cytokines then promote angiogenesis and recruitment of monocytes, creating an inflammatory milieu in the tumour environment. Furthermore, extracellular S100A4 also decreases the expression of occluding and vascular endothelial cadherin (VE)-cadherin in endothelial cells (ECs), thereby disrupting cell-cell adhesion and enabling the tumour cells to transmigrate through the EC monolayer into the bloodstream. The release of S100A8/S100A9 can be induced by UV radiation-exposed keratinocytes, and extracellular S100A8/S100A9 then promotes proliferation and migration of melanocytes via RAGE signalling. The interaction between S100A8/S100A9 and RAGE can lead to increased levels of the metalloproteinases MMP2, MMP9, and MMP14 in melanoma cells, thereby enhancing metastatic properties. In addition to RAGE, S100A8/A9 also binds to the melanoma cell adhesion molecule (MCAM), thereby activating tumour progression locus 2 (TPL2) and stimulating the transcription factor ETS translocation variant 4 (ETV4), leading to the induction of MMP25 and promoting melanoma lung metastasis. The homodimer of S100A9 additionally interacts with extracellular matrix metalloprotease inducer (EMMPRIN), which activates TNF receptor-associated factor (TRAF2)-dependent NF-κB signalling and the upregulation of cytokines such as TNFα, chemokine (C-X-C motif) ligand 1 (CXCL1), CXCL2, and CXCL3, resulting in metastasis. S100B inhibits the phosphorylation of ribosomal S6 kinase (RSK) by extracellular signal-regulated kinase (ERK) in a Ca 2+ -dependent manner so that RSK remains in the cytoplasm, leading to improved tumour survival. S100B also inhibits p53 activity at the protein level, thereby preventing p53-dependent apoptosis.

S100 Proteins in Ovarian Cancer
Recent studies investigating mRNA expression of S100 proteins in correlation with the survival of ovarian cancer patients demonstrated that high levels of the S100 family members S100A2, S100A10, S100A11, S100A15, S100A16, and S100B predict worse OS, while S100A1, S100A3, S100A5, S100A6, S100A13, S100G, and S100Z are associated with longer OS, partially depending on cancer subtype and clinicopathological features [144,145]. However, the underlying mechanisms of these S100 proteins in the context of ovarian cancer have not yet been fully understood.
Nevertheless, among others, S100A3, S100A10, and S100B were identified to be related to drug resistance in ovarian cancer [146][147][148]. Elevated expression of S100A3 was found primarily in ovarian cancer cells resistant to cisplatin and paclitaxel, both standard chemotherapeutic drugs, as well as topotecan, a chemotherapeutic agent commonly used in second-line treatment [146]. S100A10 was 1 of 11 genes associated with multidrug resistance in ovarian primary serous carcinoma tissue, although further studies are needed to validate these data, as this relation could not be confirmed in a group with a smaller sample size [147]. In vitro studies could demonstrate that S100B contributes to cisplatin resistance in ovarian cancer stem cells (OCSC) by inhibiting p53 and therefore promoting multidrug resistance gene 1 (MDR1) and MDR-associated protein 1 (MRP1) [148,149].
Moreover, it was shown that ovarian cancer patients with high expression of cytoplasmic S100A10 and stromal Annexin A2 have a 3.4-fold increased risk of progression and a 7.9-fold higher risk of cancer-related death [150]. Annexin A2 and S100A10 form a heterotetramer referred to as AIIt, which interacts with the tissue plasminogen activator (tPA). This leads to the production of plasmin, which results in ECM degradation, EMT, and angiogenesis [151,152]. In addition, S100A10 is able to promote MMPs, altogether provoking cell migration and invasion [153].

S100 Proteins in Colorectal Cancer
Mounting evidence suggests that S100 proteins play a crucial role in colorectal carcinogenesis, through promoting proliferation, migration, and invasion [154]. Several S100 family members seem to be promising biomarkers, for example, high expression of S100A4 is associated with poor survival and increased aggressiveness when studying colorectal carcinoma (CRC) patients [155,156]. S100A8/A9 could be a reliable marker for diagnosis, and postoperative monitoring of CRC [157] and S100B can predict early relapse in stages II and III postoperative colon cancer patients [158].
Regarding the role of S100A4 in colorectal cancer, it comes as no surprise that this protein seems to be related to metastasis, as similar observations were made in other cancer types, such as breast and lung cancer [90,115,159]. It was shown that extracellular S100A4 promotes ERK and hypoxia signalling in a RAGE-dependent manner, mediating colon cancer cell motility [160]. Further, S100A4 is capable of activating the phosphoinositide 3-kinase/Akt/mechanistic target of the rapamycin pathway (PI3K/Akt/mTOR), leading to increased levels of VEGF and decreased E-cadherin, which triggers tumour progression and cell migration [161].
Furthermore, high protein expressions of S100A8 and S100A9 were found to be associated with differentiation and lymph node metastasis in CRC tissue. This overexpression correlates with increased levels of β-catenin [162]. Additionally, S100A8/A9 interaction with RAGE activates NF-κB and MAPK signalling, leading to upregulation of chemokines, such as CXCL1, CLC5, and CLC7, resulting in the formation of a premetastatic niche and tumour progression [163].
Recent studies also unravelled the involvement of S100P in tumorigenesis, metastasis, and invasion of CRC [164]. In this context, thioredoxin-1 (Trx-1) and β-tubulin were introduced as novel downstream targets of S100P, and interaction between these proteins led to improved cell viability and promotion of cell migration [165,166].
2.6. S100 Proteins in Pancreatic Cancer S100 proteins were shown to be involved in tumour progression and metastasis of pancreatic cancer [164,167]. S100A2, for example, is overexpressed in pancreatic cancer tissue and seems to be a good predictor regarding the response to pancreatectomy of pancreatic cancer patients [168,169]. S100A6 was found to be overexpressed, especially in the early phase of carcinogenesis, and high nuclear S100A6 levels predict poor survival of pancreatic cancer patients [170,171]. Similar observations were made for S100A11, as the expression is detected in the earlier stages and seems to decrease during tumour progression [172]. Besides, it has been suggested that the expression of S100A11 correlates with poor prognosis of pancreatic cancer patients after surgical resection [173]. Furthermore, evidence indicated that S100P could be a helpful marker for the diagnosis of pancreatic cancer, as it is specifically expressed in cancer cells [174,175].
Several studies also introduced S100A4 as a potential biomarker, as its expression correlates with malignancy, metastasis, invasion, and overall poor prognosis in pancreatic cancer [176][177][178]. Furthermore, it was found that S100A4 is overexpressed in pancreatic adenocarcinoma, most likely due to hypomethylation at intron 1 of the S100A4 gene [179]. S100A4 promoted cell growth and cell motility and inhibited apoptosis in vitro [180,181]. The knock-down of S100A4 did not only induce apoptosis but also increased gemcitabine sensitivity [182]. By studying the underlying mechanisms of S100A4 in pancreatic cancer, it could be shown that S100A4 is upregulated by sonic hedgehog (Shh)-Gli-1 signalling as part of the hedgehog pathway, eventually resulting in EMT and further leading to metastasis [183]. Recent in vivo studies, including an orthotopic human pancreatic cancer xenograft mouse model, revealed that S100A4 activates focal adhesion kinase (FAK) and Src kinase, both involved in tumour progression and metastasis [184].
High expression of S100A8 and S100A9 was found in the stroma cells, mainly monocytes, of pancreatic cancer, and the number of S100A8 positive stroma cells correlated with the Smad4 status of the tumour. Smad4 is a key mediator for TGFβ, a pro-apoptotic protein involved in pancreatic tumour progression, and tumours without Smad4 expression showed a decreased number of S100A8+ stroma cells [185,186]. However, further studies are needed to fully understand the underlying mechanisms.

S100 Proteins in Other Cancer Types
A link between S100 proteins and additional cancer types, such as prostate cancer, bladder cancer, liver cancer, and oral cancer, could be made, although these relations have not yet been examined as closely as in previously described cancer types. Nevertheless, the current knowledge of S100 family members in these cancer diseases will be summarised in the following chapter.
In prostate cancer, S100A8 and S100A9 are upregulated and might be helpful markers in the early stage of prostate tumour progression [187]. In vitro studies showed that the heterodimer S100A8/A9 is secreted by prostate cancer cells, and extracellular S100A8/A9 activates NF-κB and MAPK signalling [188]. In contrast, S100A6 is present in prostate basal cells but absent in malignant prostate cells, indicating that S100A6 could expand the current spectrum of biomarkers for diagnosis [189]. However, more studies are needed to validate these findings and to further understand the role of S100 proteins in prostate cancer.
PCR-based analysis of human bladder cancer tissue showed that mRNA levels of S100A2, S100A3, S100A5, S1007, S100A8, S100A9, S100A14, S100A15, S100A16, and S100P were elevated, while S100A1, S100A4, and S100B were downregulated compared with normal bladder urothelium [190]. However, it is important to notice that only small sample size (n = 10) was examined and that a repeat of this study with a greater sample size could improve statistical stability. Nevertheless, the expression of S100A8 in bladder cancer was associated with disease progression in non-muscle-invasive bladder cancer (NMIBC) in more recent studies [191,192]. Investigating S100 proteins in bladder cancer also revealed a correlation between S100P and drug resistance, as decreased S100P levels were found in cisplatin-resistant bladder cancer cells, whereas overexpression of S100P increased sensitivity to cisplatin [193].
The role of S100 proteins in hepatocellular carcinoma (HCC), the most common form of primary liver cancer [194], is poorly understood so far. However, it was shown that high expression of S100A4 in association with high expression of vimentin and low expression of E-cadherin correlates with decreased OS [195]. Moreover, in vivo mouse model studies could demonstrate that the depletion of S100A4+ stroma cells, including mainly fibroblasts and macrophages, within the TME of HCC, reduced stemness of the tumour, and inflammation [196]. However, Li et al. presented evidence showing that S100A4 in cooperation with collagen I is involved in the development of fibrosis-associated HCC by upregulating stem cell marker genes, for example, Oct-4, Nanog, and SOX2, via RAGE-dependent β-catenin signalling [197]. Recently, S100A14 was introduced as a promising marker, as it was overexpressed in HCC tissue compared with healthy liver tissue and cirrhosis tissue, and therefore might be a sensitive assistant for diagnosis of HCC [198,199].
The expression of S100 proteins in oral squamous cell cancer (OSCC) remains controversial, as expression profiles vary from study to study, and further investigation is needed to identify reliable biomarkers among the S100 family members in this context [200]. Nevertheless, S100A14 seems to have tumour-suppressing properties in oral cancer, as overexpression of S100A14 in OSCC cell lines resulted in suppression of invasion, associated with the downregulation of MMP1 and MMP9, as well as decreased proliferation, correlating with upregulation of p21 and G 1 cell cycle arrest [201,202].
The evaluated expression of S100 proteins could also be detected in malignant mesothelioma [203]. In this context, S100A4 was shown to be a promising biomarker, as it correlated with tumour progression, morphologic changes related to EMT, invasion, and metastasis [204]. However, more research is needed to validate the role of S100 family members in malignant mesothelioma.
Soluble factors originating from glioma cells have been linked to affecting the tolerance of microglia, which function as mediators of innate and adaptive immune responses. The activation of immunosuppressive signalling pathways such as STAT3 blocks the effector function of microglia [205][206][207][208]. There is growing evidence that the RAGE ligand S100B is another factor that plays a role in glioma progression. Low levels of S100B expressed by gliomas triggered STAT3 and blocked microglia and also macrophage activation [209]. S100B may also encourage glioma growth by TAM chemoattraction and therefore, infiltration into gliomas through the upregulation of CCL2 [210]. In a high-throughput screening cell-based S100B promoter-driven luciferase reporter assay, duloxetine was identified to inhibit S100B and CCL2 production in a mouse glioma model. The inhibitor had the potential to modulate the immune-suppressive behaviour of TAMs [211]. Moreover, it has been revealed that high S100B levels correlate with poor prognosis in recurrent glioma patients [212].

Targeting S100 Proteins for Cancer Therapy
As it is now well established that S100 proteins are widely involved in carcinogenesis, invasion, and metastasis, a broad spectrum of novel therapeutic opportunities arises. There are several different strategies for targeting the S100 family members in cancer. While some inhibitors seem to be effective by inhibiting the transcription of S100 genes, others inhibit S100 protein activity by disturbing the interaction between S100 proteins and their targets. In addition, targeting covalent modifications, such as S-nitrosylation, S-glutathionylation, and phosphorylation, could be a promising strategy, as these modifications influence the function of S100 proteins [213]. In this review, S100 inhibitors are classified into small molecule inhibitors, neutralising antibodies, and microRNA (miRNA) mimics, and a summary is given in Table 1. Table 1. Overview of potential S100 inhibitors.

S100A8 silencing leads to decreased proliferation and invasion of laryngeal carcinoma cells and increases sensitivity of endometrial carcinoma cells for paclitaxel
Pre-clinical [242,243] miR-6884-5p S100A16 S100A16 silencing reduces proliferation, invasion, and EMT in gastric cancer Pre-clinical [244] 3.

Small Molecule Inhibitors
Small molecule inhibitors are compounds (<500 Da) able to penetrate tissue more effectively than macromolecules and are usually suitable for oral administration [245]. Although several small molecule inhibitors have been approved for cancer treatment, it is essential to further identify new compounds, to overcome drug resistance [246].
In this context, small molecule inhibitors were found which inhibit the transcription of S100 genes. The expression of S100A4, for example, is mediated by Wnt/β-catenin signalling, a pathway which is highly involved in many cancer types, especially colon cancer. Therefore, inhibitors were introduced to inhibit the transcription of the S100A4 gene by interfering with the Wnt pathway [45]. Calcimycin, a calcium ionophore, and sulindac sulfide (sulindac), a nonsteroidal anti-inflammatory drug, inhibit the expression of β-catenin, leading to reduced levels of target genes, including S100A4. Treatment of mice with these inhibitors resulted in decreased tumour growth, reduced invasion, and fewer metastases of colon cancer at least partially due to lower levels of S100A4 [214,215]. Furthermore, the U.S. Food and Drug Administration (FDA)-approved anthelmintic drug niclosamide was identified to inhibit S100A4-induced metastasis formation of colon cancer, by preventing β-catenin/T-cell factor (TCF) complex formation and therefore transcription of S100A4 [216]. Niclosamide tablets are currently being studied in a phase II clinical trial for the treatment of metastasised CRC [217]. Although these inhibitors seem to be very effective in pre-clinical studies, it is vital to notice that they are not selective inhibitors of S100A4, for example, the inhibition of β-catenin also results in downregulation of cyclin D1 and the proto-oncogene c-Myc [214], which might cause unfavourable side effects [247].
Other small molecule inhibitors, therefore, follow the strategy of inhibiting the target interaction of S100 proteins. Trifluoperazine (TFP), a phenothiazine, for example, was introduced as an S100A4 inhibitor, as TFP and S100A4 form heterodimers and five dimers then arrange to a pentameric ring. This sequestering mechanism inhibits the interaction between S100A4 and myosin-IIA, indicating that TFP might influence cell motility [218,219]. Structure-based virtual inhibitor screening could also identify substituted 1,2,4-triazoles as potential S100A10 inhibitors, as it was demonstrated that 1,2,4-triazole competes with annexin A2 over the binding to S100A10 [224,225]. However, the effects of TFP and 1,2,4-triazoles in vitro and in vivo have yet to be demonstrated.
Furthermore, tasquinimod, a quinoline-3-carboxamide derivative, could be identified as an S100A9 inhibitor. Tasquinimod binds to S100A9 and therefore blocks the interaction with RAGE and TLR4, which influences the TME by modulating myeloid cell populations [220][221][222]. Recently, a phase III clinical trial was completed, investigating the effect of tasquinimod on patients with metastatic castration-resistant prostate cancer (mCRPC). This study showed that treatment with tasquinimod resulted in significantly longer radiologic progression-free survival compared with the placebo group (7.0 months vs. 4.4 months), but no influence on the OS was observed (24 months for placebo vs. 21.3 months for tasquinimod treatment) [223].
As previously described, the interaction of S100B and p53 is a promising target, especially in melanoma [54,142,143]. In this context, pentamidine, an antiprotozoal agent, was identified to bind to the p53 binding site of S100B, thereby inhibiting S100B/p53 interaction, allowing p53 to restore tumour-suppressing properties [226,227]. Pentamidine is currently being evaluated within a phase II clinical study for the treatment of refractory wild-type p53 melanoma [228].
Another noteworthy S100 inhibitor is the anti-allergy drug cromolyn. Cromolyn binds to S100P, thereby inhibiting interaction with RAGE, leading to decreased proliferation, invasion, and NF-κB activity in vitro and reduced tumour growth in vivo. Moreover, cromolyn increased gemcitabine sensitivity for pancreatic cancer models [229,230]. However, high concentrations of cromolyn were needed to achieve these effects. A more recent study showed that the analogue 5-methyl cromolyn is more efficient than cromolyn, and lower doses could induce similar reactions [231].

Neutralising Antibodies
Therapeutic antibodies have brought great benefit to cancer therapy, as they are more selective compared with conventional chemotherapeutics, and evolving technology has enabled the development of 30 FDA-approved monoclonal antibodies for cancer treatment so far [248,249]. The anti-cancer effect of these antibodies is either achieved by immune-mediating mechanisms or by neutralising important players of tumorigenic pathways [250,251]. The latter strategy seems to be an exciting approach to target S100 proteins.
As it was shown that S100A4 is highly involved in metastatic processes through influencing the TME, it is a favourable strategy to neutralise S100A4 activity. In this context, an anti-S100A4 antibody, named 6B12, showed immunomodulatory activity by binding S100A4 and thereby preventing T-cell attraction to the tumour side, which reduced metastasis in lung and breast cancer within experimental mouse models [232,233]. The anti-S100A4 antibody 5C3 could show similar effects as it terminated endothelial tumour growth, cell migration and angiogenesis in vitro and in vivo for pancreatic cancer and melanoma models. In addition, a synergistic effect between 5C3 and gemcitabine was observed [234].
Furthermore, a monoclonal antibody targeting extracellular S100A7 was designed. It could be demonstrated that this anti-S100A7 antibody, named 6F5, blocks S100A7/RAGE interaction, thereby inhibiting S100A7-mediated MMP9 activity, leading to decreased tumour growth, cell migration, and angiogenesis in a xenograft cancer model [235]. In terms of neutralising the DAMP molecules S100A8 and S100A9, monoclonal Ab45 was identified to be the most selective among ten anti-S100A8/S100A9 antibodies. In vitro, Ab45 could inhibit S100A8/S100A9-stimulated chemotaxis, and in vitro significantly reduced lung metastasis within a melanoma model [236]. The anti-S100P antibody 2H8 was introduced to inhibit S100P-induced proliferation of pancreatic cancer cells and blocked increased survival induced by S100P after gemcitabine treatment. Moreover, it was suggested that 2H8 is more effective than the small molecule inhibitor cromolyn, as cromolyn has a more significant impact on only tumour growth, while 2H8 reduced tumour growth and additionally inhibited liver metastasis in vivo [237].
However, the development of S100 antibodies for cancer therapy is at its very beginning, and examples of other therapeutic antibodies showed that limited clinical efficacy, for example, due to lacking tumour tissue penetration, immune reactions, and antibody resistance [252], is an issue which might need to be overcome in the later stages of this process.

Micro RNA (miRNA) Mimics
MiRNAs are short, non-coding nucleotides, which regulate gene expression on the post-transcriptional level by binding to their target mRNA and thereby either repress translation or tag the mRNA for degradation [253]. In recent years, new technology was established, which exploits this natural process of gene silencing by using miRNA mimics, chemically synthesised double-stranded RNAs, which act as mature miRNA once transfected into the cell [254].
Several miRNAs were introduced to target the expression of S100 proteins. Among them, two miRNAs, namely miR-187-3p and miR-149-3p, were found to downregulate S100A4 expression. The treatment of HCC in vitro and in vivo with miR-187-3p resulted in reduced metastasis and EMT, while miR-149-3p could inhibit invasion and migration of bladder cancer cells [238,239]. Furthermore, miR-193a suppressed proliferation, invasion, migration, and angiogenesis in vitro and in vivo within lung cancer models by silencing S100A6 [240].
The expression of S100A7 could be downregulated by miR-26b-5p, leading to decreased proliferation, migration, and invasion of intrahepatic cholangiocarcinoma in vitro [241]. Similar effects were observed for miR-24, a miRNA targeting S100A8, which inhibited proliferation and invasion of laryngeal carcinoma cells [242]. Besides, it was shown that S100A8-silencing by miR-24 reduced proliferation but enhanced sensitivity to paclitaxel of endometrial carcinoma cells [243]. Another interesting approach demonstrated that miR-6884-5p targets the expression of S100A16, which resulted in reduced proliferation, EMT, and invasion of gastric cancer cells in vitro [244].
The use of miRNA mimics, however, is a relatively new strategy for cancer therapy, and appropriate delivery systems need to be optimised to achieve effective treatment in the clinical context [255].

Conclusions
Research over the past decades could show that S100 proteins are highly involved in many different cellular processes and are essential players in pathophysiological mechanisms. S100 family members are widely expressed proteins in vertebrates, yet their way of functioning is strongly dependent on the tissue context. Several exciting and promising approaches were made to exploit the current knowledge and use S100 proteins as valuable associates in the context of cancer therapy. However, far more research is needed to broadly establish S100 proteins as reliable biomarkers and to identify and further optimise safe and efficient S100 therapeutics. Given initial promising results, a better understanding of S100 proteins and their properties will bring great benefit to novel clinical applications.    Modulation of the cytoskeleton: interacts with several proteins of the cytoskeleton (e.g., keratin and F-actin) and promotes polymerisation of microtubules Protection against pathogens: S100A8/S100A9 expressing epithelial cells are more resistant to bacterial infection Inflammation: RAGE-and TLR4-mediated upregulation of pro-inflammatory cytokines (e.g., IL-6 and IL-8) and adhesion proteins (e.g., ICAM-1) [29] S100A10 (p11) Endothelial cells, macrophages, fibroblasts, epithelial cells The expression can be induced by: Sp1, IFN-γ, glucocorticoids, TGFβ, EGF, IL-1β, thrombin, oncogenes (e.g., PML-RARα, KRas) Annexin II, 5-HT1B receptor, TRPV5, TRPV6, TASK-1 Role as plasminogen receptor: regulates plasmin production by forming a complex with annexin II, which leads to fibrinolysis, activation of MMPs, ECM degradation, stimulation of the JAK1/TYK2 signalling pathway, and recruitment of macrophages Trafficking of plasma membrane proteins: S100A10 interacts with the 5-HT1B receptor, TRPV5, TRPV6, and TASK-1 and participates in their trafficking Cell growth regulation: intracellular S100A11 inhibits cell growth by binding to nucleolin, which allows expression of p21, leading to inhibition of DNA synthesis; extracellular S100A11 promotes cell proliferation by RAGE-mediated transcription of EGF Interaction with RAGE: RAGE activates p38 MAPK kinase, which increases type X collagen [49,50]  Non-canonical secretion pathway: forms a copper-dependent multiprotein complex with Syt1 and SPHK and is thereby involved in the non-classical stress-dependent release of FGF-1, IL-1α, and prothymosin-α [265][266][267]

S100A14
High expression in epithelial tissue, lower expression in mesenchymal tissue Regulated by p53 RAGE, p53 Interaction with RAGE: binding in lower concentrations activates MAPK and NF-κB signalling, leading to cell proliferation; binding in high concentrations activates ROS production, resulting in apoptosis [46,53] S100A15 (S100A7A; koebnerisin) Dendric cells, endothelial cells, vascular smooth muscle cells, peripheral nerves, keratinocytes Upregulated by E. coli through TLR4 Transcription can be induced by IFN-γ, IL-1β, TNFα, and Th1 Co-expression with S100A7 in keratinocytes GPCR Epidermal cell maturation: upregulated in epidermal differentiation Initial immune response: functions as an antibacterial agent by reducing the survival of E.
coli and other strains Inflammation: acts as a chemoattractant for leucocytes via GPCR [23,268] S100A16 Astrocytes, adipocytes Ca 2+ influences nuclear import/export p53 Overexpression in preadipocytes resulted in increased proliferation and reduction in insulin-stimulated glucose uptake and Akt phosphorylation Promoter has binding sites for SMAD, STAT/CREB, and SP/KLF During embryonic implantation: highly expressed in the trophoblastic layer of the embryo, and in the endometrium of the uterine wall RAGE, IQGAP1, enzrin, NMIIA Cell proliferation and survival: activates RAGE-mediated NF-κB signalling; interaction with IQGAP1 induces MAPK signalling cascade Cell migration: promotes interaction with F-actin (via enzrin) and reduces focal adhesion sites (via NMIIA) [48,272] S100Z Leucocytes --- [8]