Starts 153516089 bp from pter; ends 153522612 bp from pter; 6524 bases; orientation minus strand.
Homo sapiens chromosome 1, GRCh37 primary reference assembly.
NCBI reference sequence: NC_000001.10, NT_004487.19.

The human S100A4 gene has four exons. Exon 1 is non-coding and exons 2 and 3 are coding exons. Exon 2 with the start codon and encodes N-terminal EF hand and exon 3 encodes the C-terminal EF-hand. The fourth non-coding exon occurs in the 5-UTR.

Two variant RNA transcripts from Source Search. The NM_019554 is a longer transcript. NM_002961 possesses alternate 5-UTR; both encode the same protein isoform.
NCBI reference sequence NM_019554 ver 01954.2 564 bp mRNA; NM 002961 ver 002961.2; 512 bp mRNA.
Splice variants have been reported of S100A4. In human osteosarcoma, alternative splicing within the 5-untranslated region (UTR) generates the two variants. Both variants, hu-mts1 and hu-mts1 (var), contain one open reading frame, differ slightly in translational capacity; possess similar stability. The hu-mts1 and hu-mts1 (var) splice variants with exons 1, 2, 3, and 4 and another with exons 1, 3 and 4, may be differentially expressed. The hu-mts1 (var) is expressed in the colon but not in the liver; it was not found in leukocytes, neutrophils, macrophages and lymphocytes. The hu-mts1 variant predominated in human breast carcinoma (SK-BR-3) and lung carcinoma (A549) is predominant (Ambartsumian et al., 1995). The differential association of the variants has been described in gastric cancers, which seems to relate to disease state possibly relates to progression; however the expression status of the variants in the lymph node metastases is not known. A splice transcript with loss of non-coding exon 1a/1b, but exons 2 and 3 present has been described in infiltrating carcinoma of the breast by Albertazzi et al. One would note nonetheless that Alternative Splicing and Transcript Diversity (ASTD) have listed 12 variant transcripts.

Regulation of transcription
Binding sites for several transcription factors have been identified in the promoter of S100A4. SABiosciences ChIP-qPCR Assay database lists 19 p53 binding sites.
Multiple NFAT (nuclear factor of activated T cells) transcription factor consensus binding sites; NF-kappaB related binding site (Tulchinsky et al., 1997). Much evidence is also available regarding activation of NF-kappaB by S100A4. S100A4 can activate NF-kappaB via the classical pathway mediated by MEKK/IKKβ; S100A6 and S100P also are capable of exerting pro-metastatic effects again by activating the NF-kappaB pathway. Experimentally induced expression of S100A4 is inhibited by NF-kappaB inhibitors. Aside from these, several other regulatory pathways may be identified, e.g. the Wnt/β-catenin/TCF, HIF/HER among others, as evidenced by the established phenotypic expression induced by the gene.
S100A4 has been postulated to signal via RAGE (receptor for advanced glycation end products) which is known to activate NF-kappaB.
A composite enhancer consisting of 6 cis-elements has been identified in the first intron of murine S100A4. This interacts with Sp1 and AP-1 family members and CBF (core binding factor alpha) and KRC (zinc finger transcription factor kappa recognition component) transcription factors.

None reported.

Human S100A4 (also mouse and rat S100A4) contains 101 aminoacid residues and is approx 12 kDa in size. In common with most S100 family members, S100A4 is an antiparallel homodimer stabilised by noncovalent interactions between two helices from each subunit forming an X-type four-helix bundle. Each subunit has two calcium-binding EF-hands linked by the intermediate hinge region and a distinctive C-terminal extension. A pseudo-EF hand formed by helices 1 and 2 and the pseudo-EF-hand and a canonical EF-hand that are brought into proximity by a small two-stranded antiparallel beta-sheet. The hinge region and the C-terminal loop of S100 proteins are involved in target protein binding. Calcium binding produces a conformational change, which leads to the exposure of hydrophobic pocket of residues in helices 3 and 5, the hinge region and the C-terminal loop. This conformational change is required for target protein binding. S100A4 might be post-translationally modified. Charged variants conceivably resulting from post-translational changes have been described in one report, but no confirmation of these findings has been forthcoming. However, calculations from the predicted isoelectric point of S100A4 and separation of the charged variants from the major spot would suggest that two variants may have displayed 17.4 and 26.1% and a third variant with a possible highly extended form and nearly 2.6-fold increase in net negative charge. Alterations in net molecular charge of this magnitude and charge distribution can alter protein configuration.
NCBI sequence: NM_002961; NP_002952; NM_019554; NP_062427; UniProtKB/Swiss-Prot: P26447.

Features
EF-hand domains:
EF hand 1: length 36; position 12-47,
EF hand 2: length 36; position 50-84.
Target protein interaction domains: in the active state S100A4 interacts with many target proteins e.g. p53 family proteins, HDM2, Annexin II, F-actin, tropomyosin, and heavy chain of non-muscle myosin IIA, among others. In a closed conformational state S100A4 is inactive, but the protein assumes an open conformation upon calcium binding. In the altered configuration S100A4 can interact with target proteins. These target proteins interact with specific binding domains of S100A4, which are accessible upon conformational change of the apoprotein upon Ca²⁺ binding. The Rudland/Barraclough group has shown that specific mutations that inhibit self-association of S100A4 markedly reduce its metastasis promoting effects. The mutations reduce self-association and reduce the affinity of S100A4 to two target proteins viz. p53 and non-muscle myosin heavy chain isoform A. The interaction between S100A4 and target proteins can possibly also be disrupted by the packaging of S100A4 in such a way as to sequester S100A4 dimers.
Inhibition of S100A4 polymerisation by suppressing TG2 (tissue transglutaminase 2) function has resulted in the inhibition of cell migration in vitro. This is inspired by the fact that TG2 is a cross-linking protein. Treatment of cells in vitro with EGF seems to up regulate the expression of EGFR and TG2 accompanied by enhanced cell migration. S100A4 over expressing tumours not infrequently tend to be EGFR postive; so tissue transglutaminase could be promoting EGFR dimerisation and facilitate EGF/EGFR signalling.

S100A4 is distributed ubiquitously in normal tissues (Mazzucchelli, 2002).
For expression profile: Human Protein Atlas (HPA): CAB002618 and Human Protein Reference Database HPRD.
S100A4 occurs in many forms of human cancer, e.g. breast, colorectal, liver, lung, head and neck, ovarian, endometrial, pancreatic, renal, testicular, and prostate cancers, and melanoma; also in many cell lines of myeloid, lymphoid, lung and brain origin and cell lines derived from many forms of leukaemias. The expression of the gene is regulated by methylation. Over expression correlates with hypomethylation and the frequency of hypomethylation relates to tumour progression, e.g. in ovarian cancers. There is no implication at present that the degree of methylation is related to expression. S100A4 has been implicated in other human diseases, e.g. Crohns disease and rheumatoid arthritis.

S100A4 occurs extracellulary and also in cytoplasmic and nuclear location. Differential distribution has been reported between stromal components of primary and metastatic tumour. Patterns of distribution could vary between tissues and between species. No firm functional link has been made with the site/s of localisation.
Intracellular distribution is an important factor in determining genetic activity.
It may be noted here that S100A4 is often expressed in component inflammatory cells of tumour stroma. It has been postulated that interactions between the stroma and tumour cells lead to the expression of the protein and modulate its function in either or both. However, both the postulate and its potential influence in tumour progression are yet to be established.
The pattern of intracellular distribution of many genetic determinants has been found to be highly relevant to invasion and metastasis. Nuclear location of S100A4 was shown some while ago to relate to aggressive tumour behaviour and poor prognosis. Translocation to the nucleus has been associated with EMT induced by TGF-β/Smad signalling. IL-induced translocation seems to require sumoylation of specific lysine residues and in this way conceivably regulating target gene expression. Expression patterns need to be explored in more than one tumour system. This might be crucial in the development of strategies of treatment targeting S100A4, especially with the postulated link of S100A4 expression with chemoresistance.

S100A4 protein promotes metastasis, functions as a counter point to metastasis suppressor nm23, and is implicated in the regulation of the cell cycle, cell proliferation, motility, invasion, tubulin polymerisation, and angiogenesis. S100A4 might suppress expression of other suppressor genes e.g. PRDM2 and VASH1. PRDM2 (PR domain containing 2, with ZNF domain) is a tumour suppressor gene encoding a zinc finger protein. VASH1 (vasohibin 1) inhibits cell migration, proliferation and tumour growth and angiogenesis.
S100A4 promotes metastatic spread of cancer as demonstrated by gene transfer studies. Its expression has shown clear correlation with tumour spread to lymph nodes and with prognosis.

Cell cycle, cell proliferation, tumour growth and apoptosis.
S100A4 binds to and forms complexes with p53 to regulate cell cycle progression. P53 has been confirmed as a target of S100A4, which stabilises p53. There is conclusive evidence that S100A4 binds to C-terminal regulatory region of p53. S100A4 and certain other members of the S100 family bind to TAD transactivation domain (residues 1-57) of p53. They may also affect p53 function by binding to the tetramerization domain of p53 (residues 325-355) and interfering with intracellular translocation and subcellular localisation. This interaction is suggested to be linked with p53 function. Nineteen p53 binding sites have been identified in the promoter of S100A4 (SABiosciences ChIP-qPCR Assay). S100A4 also influences p21^waf1 and mdm2, a regulator of p53 function and the apoptosis family bax gene. It binds to N-terminal domain of mdm2. Signalling pathways include P53-Rb/stathmin/p53 down stream effectors, e.g. p21^waf, p16 etc. P53/stathmin signalling modulates microtubule dynamics and cell division. Furthermore, p53 and down stream target apoptosis family genes such as BNIP3, caspases; calpain/Fas (?) are postulated as important pathways in S100A4 signalling. Knockdown of S100A4 has been reported to lead to apoptosis. The transcription factor NF-kappaB which involved in anti-apoptosis has been implicated in S100A4 signalling.
S100A4 proliferative signalling seems to involve epidermal growth factor receptors (EGFR). EGFR expression correlates with S100A4 expression. Interactive signalling with HER2 might be postulated with the finding that S100A4 stimulates EGFR/HER2 receptor signalling and on the identification in human S100A4 promoter of an HER2 response element 1099-1487 bp up stream of the transcription start site. The interaction of S100A4 with the TGF-beta system via Smad has also been reported. S100A4 seems able to bind to the N-ter region of Smad3. TGF-beta is an important activator of epithelial mesenchymal transition leading to acquisition of invasive ability. The interaction between S100A4 and Smad thus falls in place with the metastasis-promoting function of the former. Some of these pathways are pictorially represented above (figure 5). S100A4 activates EMT via up regulation of Snail, a negative regulator of E-cadherin. The TGF-β family receptor Activin involvement has been implicated.

Invasion, motility, and intercellular adhesion.
One of the targets of S100A4 involved in cell motility is myosin filaments. Myosin II consists of two heavy chains (MHC) with globular domains which interact with F-actin. The tail domains of heavy chains form a coiled-coil tail that participates in the assembly of myosin filaments. Wrapped round the neck region of each heavy chain are the essential and the regulatory light chains. Phosphorylation of the regulatory light chain and also of MHC plays an important part in the assesmbly of myosin II monomers into filaments. S100A4 inhibits CK2-mediated phosphorylation of MHC, inhibits the assembly of myosin monomers into filaments. The affinity of S100A4 for the myosin-IIA can be reduced by CK2-mediated phosphorylation. S100A4 destabilises MHCIIA filaments phosphorylated by PKC and inhibits the assembly of monomers. PKC and CK2 can phosphorylate distinct serine residues but yet be additive in their effect. The outcome is that S100A4 promotes dissociation of the filaments and prevents self assembly of monomers resulting in enhanced migration. Thus S100A4 seems to provide a mechanistic link between the actomyosin cytoskeletal and migration.
Signalling systems include modulation of cytoskeletal dynamics; cadherin/catenin complex cytoskeletal linkage and significantly a TCF, a component of the canonical Wnt signalling system, binding site has been identified in the S100A4 promoter and S100A4 directly binds heterodimeric beta-catenin/TCF complexes; CD44/cytoskeletal linkage; ECM associated proteolytic enzyme system/ECM remodelling, affects tubulin polymerisation. S100A4 and tumour suppressor nm23 exert opposite effects on tubulin dynamics. Two C-terminal lysine residues are required for enhanced motility and invasion and interaction with target proteins. The connective tissue growth factor (CTGF) has been reported to up regulate S100A4 expression and inhibition of S100A4 blocks CTGF-induced cell motility.
S100A4 seems to function via the MMP/TIMP system in promoting invasion as well as induction of angiogenesis. S100A4 is over expressed in invasive glioma cell lines together with down regulation of TIMP-2, indicating a close linkup of S100A4 with the MMP system in the promotion of invasion.
Angiogenesis signalling occurs via activation of MMP/TIMP; activation of angiogenic factors VEGF/endothelial cell proliferation; MetAP2/p53-mediated inhibition of endothelial cell proliferation. S100A4 stimulates angiogenic signalling in breast cancer. An indirect link is suggested by the inhibition of S100A4 by Interferon-gamma which might inhibit angiogenesis by down regulating VEGF expression. Hypoxia is a major regulator of angiogenesis. HIF-1α (hypoxia-inducible factor-1α) is a transcription regulator in hypoxia. It can activate VEGF to induce angiogenesis and TGF-α and promotes cell survival. Exposure to hypoxia has been correlated with reduced methylation of the hypoxia response element in S100A4s promoter region and enhanced HIF binding to the promoter and increased transcription of the gene together with increased cell proliferation and invasion. Given that HIF also promotes VEGF expression one can see a potential two pronged approach to control tumour growth with HIF inhibition. Some clinical studies are underway to study the effects of Sorafenib-mediated inhibition of HIF-1α and VEGF. In laboratory studies Sorafenib has been found to reduce tumour growth and tumour associated microvessel density.
Osteopontin was identified as a metastasis-associated protein some time ago. Many strands of evidence suggest that osteopontin is an intermediary in S100A4 signalling pathway. In breast cancer expression of osteopontin in the background of S100A4 has generally correlated with poor patient survival.
Osteopontin is associated with several activated NF-kappaB pathways. S100A4 induces the expression and secretion of osteopontin in some osteosarcoma cell lines in an NF-kappaB-dependent fashion. Inhibition of osteopontin inhibits tumour development and angiogenesis; inhibition of both might result in synergistic suppression of tumour progression.
Shown below are the potential pathways of S100A4 signalling in cell motility/invasion and angiogenesis, emphasising the possibility that S100A4 seems able to influence many significant systems leading to angiogenesis.

Sequence homology to protein from Pan troglodytes (Chimpanzee) (Gene ID: 457320; Protein NCBI RefSeq: XP_001138744.1).
Bos taurus (Bovine) (Gene ID: 282343).
Canis lupus familiaris (Gene ID: 403787; NCBI reference sequence: NP_001003161.1; protein: NP_777020.1).
Sequence homology 93% to murine S100A4 (Entrez Gene ID 20198; NP_035441).
Sequence homology 91% to rat protein (Entrez Gene ID 24615; NP_036750).

Many SNPs have been identified; 7 shown in NCBI and 26 in Applied Biosystems data source. The NCBI Entrez SNP database lists 19 submissions.

Chromosomal rearrangements
The locus 1q21 is a hotspot for chromosomal rearrangements, microdeletions and duplications; significance uncertain and there are no clear implications for metastasis. No translocations leading to hybrid S100A4 have been recorded.
There are 11 common and 1 rare fragile sites on chromosome 1. The common FRA1F occurs in 1q21. Chromosome 1 is prone to sister chromatid recombination (SCR) and >70% SCRs occur at the fragile sites or in the same band as the fragile sites, but no link with S100A4 established.

None reported.

No simple mutations, gene fusions, or structural variants detected in breast and colorectal carcinomas and in gliomas (Cosmic: Catalogue Of Somatic Mutations In Cancer, Welcome Trust Sanger Institute).
No mutations have been found coding regions in human, canine and feline S100A4. Mutating phenylalanine 72 to alanine reduces functional effectiveness. Toombak (tobacco rich in tobacco-specific nitrosamine) dipping (placing between the lower lip and gums) has been indirectly linked with S100A4 mutations in oral squamous cell carcinoma, but mutations have been described also in non-dippers. The carcinoma from dippers had 4 mutations (one transition, 3 transversions) and non-snuff-dippers showed 3 mutations each (one transition, 2 transversions). The suggestion is that S100A4 mutations could be complementing the effects of more frequent mutations of p53 and p21^waf1.

There are 11 common and 1 rare fragile sites on chromosome 1. The common FRA1F occurs in 1q21. Chromosome 1 is prone to sister chromatid recombination (SCR) and >70% SCRs occur at the fragile sites or in the same band as the fragile sites, but no link with S100A4 has been established.