S100A9 belongs to the S100/calgranulin family of small non-ubiquitous cytoplasmic Ca²⁺-binding proteins of EF-hand type. The proteins were referred to "S100" because of their solubility in saturated ammonium sulphate solution. Sixteen of 21 members are localised in a cluster on human chromosome 1q21.
The clustered organization of these S100 genes is conserved during evolution (Ridinger et al., 1998). A comparison between man and mouse has shown that during evolution, the colinearity of the S100 gene cluster has been destroyed by some inversions. However, the colocalization of the myeloid expressed S100 genes such as S100A8, S100A9, and S100A12 is conserved. It has been speculated, that the structural integrity of that part of the locus is necessary for the coordinated expression of these genes (Nacken et al., 2001).
Remarkably, the S100 gene cluster is located in close proximity to a region which has been frequently rearranged in human cancer (Carlsson et al., 2005) and to the epidermal differentiation complex (EDC) (Mischke et al., 1996). EDC is a cluster of genes on chromosome 1q21 encoding proteins that fulfil important functions in terminal differentiation in the human epidermis, including filaggrin, loricrin and others. In addition, linkage analyses have identified a psoriasis susceptibility region, the PSORS4 locus, that is close to the S100 gene cluster (Hardas et al., 1996; Semprini et al., 2002). These data are important indications for the involvement of S100 genes in inflammatory as well as neoplastic disorders. It has been speculated that the rearrangements result in a deregulated expression of S100 genes associated with neoplasia.

The S100 gene structure has been structurally conserved during evolution. Similar to most S100 genes S100A9 consists of three exons that are separated by two introns.

In the S100A9 gene, exon 1 encodes the untranslated region. The protein is encoded by sequences in exon 2 and exon 3, encoding a N-terminal and a C-terminal EF-hand motif, respectively.
The sequence of human S100A9 cDNA has an open reading frame of 352 nucleotides.
S100A9 expression appears to be restricted to a specific stage of myeloid differentiation. The protein is present in circulating neutrophils and monocytes, but not in resting tissue macrophages. In peripheral blood monocytes it is down regulated during maturation to macrophages. Despite a number of distinct regulatory regions are located upstream of the transcription initiation site, the corresponding nuclear factors as well as the underlying molecular mechanisms still remain unclear. Transcription factors such as PU.1 (Henkel et al., 2002), C/EBP-alpha and C/EBP-beta (Kuruto-Niwa et al., 1998) have been shown to drive S100 gene expression within the myeloid lineage.
For example, during differentiation of HL-60 cells into monocyte-like cells two still not identified factors were found to bind to the upstream regions of S100A9 gene; one adjacent to the TATA box and another in the region between -400 and -150 (Kuwayama et al., 1993). Another study revealed a CCAAT/enhancer binding protein (C/EBP)-binding motif located at position -81 upstream of the S100A9 gene. Both C/EBP-alpha and -beta bind to this motif in a myeloid/monocytic differentiation-dependent manner (Kuruto-Niwa et al., 1998). C/EBP was shown to be alone sufficient to drive S100A9 expression in otherwise negative cells. C/EBP up-regulation is antagonized by myb, a transcription factor active in differentiated myeloid/monocytic cells (Klempt et al., 1998). The presence of distinct epithelial and myeloid-specific regulatory regions upstream of the transcription initiation site has been demonstrated by detailed deletion analysis (Klempt et al., 1999). Besides the very specific action of particular upstream DNA elements, the S100A9 gene contains a potent enhancer, which is harbored within positions 153 to 361 of its first intron (Melkonyan et al., 1998). The functional relevance of this enhancer in S100A9 expression is supported by its conservation in human and murine S100A9 genes at almost identical positions.
Promoter analyses revealed a regulatory element within the S100A9 promoter referred to as MRP regulatory element (MRE) that drives the S100A9 gene expression in a cell-specific and differentiation-dependent manner. This regulatory region is located at position -400 to -374 bp, and two distinct nuclear complexes were demonstrated to bind to this region. Interestingly, the formation of the nuclear protein complexes closely correlates with the myeloid-specific expression of the S100A9 gene and, were therefore referred to as MRE-binding complex A (MbcA) and MbcB, respectively. Analysis of one of the two nuclear complexes revealed a heterocomplex consisting of transcriptional intermediary factor 1 beta (TIF1 beta) and a yet unidentified protein with homology to KRAB domain-containing (Kruppel-related) zinc finger proteins (ZFP) (Kerkhoff et al., 2002).
Beside its expression in myeloid cells S100A9 is expressed in epithelia under specific conditions. Its expression is transiently induced in keratinocytes after epidermal injury and UVB irradiation, and the protein is expressed at extremely high levels in psoriatic keratinocytes. Furthermore, its expression is induced by pro-inflammatory cytokines such as TNF alpha and IL1 beta. Recently, a complex of Poly (ADP-ribose) polymerase (PARP-1) and Ku70/Ku80 has been demonstrated to drive the stress response-specific S100 gene expression (Grote et al., 2006). The stress response-induced expression of the S100 proteins points to an important role in skin pathology.
In breast cancer cells S100A9 gene expression is induced by the cytokine oncostatin M (OM) through the STAT3-signaling cascade (Li et al., 2004). This finding is in accordance with another study showing that IL-22 up-regulates the expression of S100A7, S100A8, and S100A9 in keratinocytes. IL-22 has been demonstrated to induce STAT3 activation in keratinocytes (Boniface et al., 2005).

Not known.

The sequence of human S100A9 cDNA has an open reading frame of 352 nucleotides predicting a protein of 114 amino acids and a calculated Mr of 13242 Da.
Beside the full-length form of S100A9 there is a truncated isoform of S100A9 resulting from alternative translation.
The full-length form of S100A9 lacks the first Met, and Thr at position 2 is acetylated leading to a calculated molecular mass of 13154 Da.
The N-truncated isoform starts with Met at position 5. Posttranslational removing of Met at position 5 and consequent acetylation of Ser at position 6 leads to a calculated molecular mass of 12690 Da.
The theoretical isoelectric point of the full length form is 5.7 and for the truncated form is 5.5, respectively.
S100A9 is composed of two helix-loop-helix EF-hand motifs. The C-terminal EF-hand contains a canonical Ca²⁺-binding loop of 12 amino acids. Conversely, the N-terminal EF-hand contains a Ca²⁺-binding loop of 14 residues that binds Ca²⁺ mostly through main-chain carbonyl groups that which is specific to S100 proteins. Consequently, S100 proteins have a weaker Ca²⁺ affinity than typical Ca²⁺ sensors such as calmodulin (Donato, 2003).
An important posttranslational modification of S100A9 represents the phosphorylation of threonine at position 113. It can be phosphorylated upon PMN activation, and phosphorylation of this residue is specifically regulated by the Ca²⁺-ionophore, ionomycin. Recent studies give evidence for S100A9 being a p38 MAPK substrate in human neutrophils (Lominadze et al., 2005). This phosphorylation is involved in translocation and functional events.
In vivo and in vitro experiments have shown that S100 proteins form homo-, hetero- and oligomeric assemblies (Hunter and Chazin, 1998; Osterloh et al., 1998; Pröpper et al., 1999; Moroz et al., 2003). Together with their specific cell- and tissue-expression patterns, the structural variations, and the different metal ion binding properties (Ca²⁺, Zn²⁺ and Cu²⁺) the S100 protein complexes might be functionally diversified. S100A9 preferentially interacts with S100A8.
It is worthwhile mentioning that the murine analogs display a stronger tendency to form homodimeric protein complexes. In view of the formation of different tertiary structures with putative distinct functions it is tempting to speculate that S100A8 and/or A9 have different functions in mouse and man.

S100A9 is mainly expressed in cells of the myeloid lineage, however, its gene expression is induced in epithelial cells in response to stress, in specific conditions such as wound healing, UV exposition, abundant in psoriais keratinocytes, differentially expressed in several cancers.

Mostly cytoplasmic, but also at membranes and cytoskeleton.
In resting phagocytes the S100A8/A9 protein complex is mainly located in the cytosol. Upon cellular activation the protein complex is either translocated to cytoskeleton and plasma membrane or released into the extracellular environment.
The translocation pathways occur upon the elevation of the intracellular calcium level (Roth et al., 1993). At a later time point, the S100A8/A9 heterodimers can be detected on the surface of monocytes (Bhardwaj et al., 1992). The mechanism by which the S100A8/A9 heterodimer penetrates the plasma membrane remains unclear since the S100 proteins lack a transmembrane signaling region.
The secretion pathway relies on the activation of protein kinase C. This pathway differs from the classical as well as the alternative secretion pathways of cytokines (Moqbel and Coughlin, 2006). It has been demonstrated that this novel secretion pathway is energy-consuming and depends on an intact microtubule network (Murao et al., 1990; Rammes et al., 1997).
Recent investigations give evidence that interaction of S100A8/A9 with annexin-6 is involved in surface expression and release of S100A8/A9 (Bode et al., 2008). Annexins are another class of Ca²⁺-regulated proteins. They are characterized by the unique architecture of their Ca²⁺-binding sites, which enables them to peripherally dock onto negatively charged membrane surfaces in their Ca²⁺-bound conformation. This property links annexins to many membrane-related events such as certain exocytic and endocytic transport steps. This is an interesting finding since S100A8 and S100A9 are expressed in cancerous cells of secretory tissues as breast and prostate. Cells originating from such glandular tissues are rich in membrane structures, suggesting that membrane-associated molecular targets for the S100A8/A9 proteins could be potentially found in these cells.
Recent investigations also demonstrated the association of S100A8/A9 with cholesterol-enriched membrane microdomains (lipid rafts) (Nacken et al., 2004). This observation is in agreement with the enhancing effect of S100A8/A9 on NADPH oxidase since the formation of the oxidase complex takes place at lipid rafts.

Intra- as well as extracellular roles have been proposed for the S100 proteins.

Intracellular activities of S100A8/A9
In the intracellular milieu, S100 proteins are considered as calcium sensors changing their conformation in response to calcium influx and then mediating calcium signals by binding to other intracellular proteins. In a mouse knock-out model chemokine-induced down regulation of the cytosolic Ca²⁺-level was detected (Nacken et al., 2005).
After calcium binding, the S100A8/A9 protein complex binds specifically polyunsaturated fatty acids. S100A8/A9 represents the exclusive arachidonic acid-binding capacity in the neutrophil cytosol (Kerkhoff et al., 1999), and participates in NADPH oxidase activation by transferring arachidonic acid to membrane-bound gp91^phox during interactions with two cytosolic oxidase activation factors, p67^phox and Rac-2. The functional relevance of S100A8/A9 in the phagocyte NADPH oxidase activation was demonstrated by the impairment of NADPH oxidase activity in neutrophil-like NB4 cells, after specifically blocking S100A9 expression, and employing bone marrow-derived PMNs from S100A9^-/- mice (Kerkhoff et al., 2005).
In accordance to their role in myeloid cells, S100A8/A9 enhances epithelial NADPH oxidases (Benedyk et al., 2007). As a consequence of enhanced ROS levels, NF-kB activation and subsequently TNF-alpha and IL-8 mRNA levels are increased in S100A8/A9-HaCaT keratinocytes, consistent with the view that NF-kB is a redox-sensitive transcription factor. Further consequences of S100A8/A9-mediated NF-kB activation are reduced cell growth, increased expression of differentiation markers, and enhanced PARP cleavage as an indicator of increased cell death (Voss et al., 2011).
In view of the stress response-induced expression of the two S100 proteins in keratinocytes these findings have great implications for tissue remodeling and repair. For example, keratinocytes acquire an activated state after cutaneous wounding in which proliferation is favored over differentiation in order to replenish the lost material and rapidly close the site of injury. Thus, it is likely to hypothesize that S100A8/A9-mediated growth reduction is required for the upcoming cell fate decision of damaged cells, i.e. for a survival phase to be followed by differentiation, proliferation, or apoptosis. These data have also an impact on tumorigenesis since S100 gene expression is associated with neoplastic disorders.
In migrating monocytes the S100A8/A9 complex has been found to be associated with cytoskeletal tubulin and to modulate transendothelial migration (Vogl et al., 2004). Investigations using two different mouse knock-out models demonstrated no obvious phenotype (Manitz et al., 2003; Hobbs et al., 2003). However, reduced migration of S100A9-deficient neutrophils and decreased surface expression of CD11b, which belongs to the integrin family, were observed upon in vitro stimulation.

Extracellular activities of S100A8/A9
The S100/calgranulins display antimicrobial activity by depriving bacterial pathogens of essential trace metals such as Zn²⁺ and Mn²⁺ (Steinbakk et al., 1990; Murthy et al., 1993; Clohessy and Golden, 1995; Sohnle et al., 2000). In the context of inflammation, it has been proposed that S100A8/A9 is massively released when neutrophils die to provide a growth-inhibitory type of host defense that is adjunctive to the usual microbicidal functions by binding metals other than Ca²⁺ (Corbin et al., 2008).
In addition, S100/calgranulins serve as leukocyte chemoattractants (Lackmann et al., 1992; Lackmann et al., 1993; Kocher et al., 1996; Lim et al., 2008). Murine S100A8 has potent chemotactic activity for neutrophils and monocytes in vitro and in vivo (Lackmann et al., 1992). In contrast, human S100A8 displays only weak leukocyte chemotactic activity in vitro and in vivo (Lackmann et al., 1993). Detailed analysis revealed that the hinge region contributes to the chemotactic activity of murine, but not human S100A8. These data questioned whether the proteins are orthologs since there is a high degree of homology between murine and human S100A8 but a functional divergence. Intriguingly, human S100A12 is chemotactic and the hinge region of human S100A12 has been implicated herein (Yang et al., 2001). Thus, the functional and sequence divergence suggested complex evolution of the S100 family in mammals.
The putative pro-inflammatory functions of S100A8 and S100A9 have recently been investigated in two different mouse knock-out models. S100A9 deficiency did not result in an obvious phenotype (Manitz et al., 2003; Hobbs et al., 2003). However, reduced migration of S100A9-deficient neutrophils and decreased surface expression of CD11b, which belongs to the integrin family, were observed upon in vitro stimulation. In addition, chemokine-induced down regulation of the cytosolic Ca²⁺-level was detected. Obviously, these in vitro effects are compensated by alternative pathways in vivo.
Remarkably, cancer cells utilize S100A8 and S100A9 as guidance for the adhesion and invasion of disseminating malignant cells (Hiratsuka et al., 2006). In the context of malignancy it was reported that S100A8/A9 attracts Mac-1⁺ myeloid cells to the lung tissue. Recruited Mac-1⁺ myeloid cells in lung in turn produce S100A8/A9 in response to primary malignant cells in a so called "premetastatic phase". This phase shows the general characteristics of an inflammation state which facilitates the micro-environmental changes required for the migration and implantation of primary tumor cells to lung tissue. After preparation of the target tissue for accepting the malignant cells, tumor cells mimic Mac-1⁺ myeloid cells in response to S100A8/A9 chemotactic signaling and migrate to lung. So, it seems that tumor cells and Mac-1⁺ myeloid cells utilize a common pathway for migration to lung which involves the activation of mitogen-activated protein kinase pathway (Hiratsuka et al., 2006). These findings suggest S100A8/A9 as an attractive target for the development of strategies counteracting tumor metastasizing to certain organs.
S100A8 and S100A9 have been identified as important endogenous damage-associated molecular pattern (DAMP) proteins. Although receptors for S100A8/A9 are still largely uncharacterized, more recent findings support the notion that they function as potent ligands of pattern-recognition receptors, such as the toll-like receptor 4 (TLR4) (Vogl et al., 2007) and the receptor for advanced glycation end products (RAGE) (Srikrishna and Freeze, 2009).
The S100/calgranulins display cytokine-like functions, including activation of the receptor for advanced glycation endproducts (RAGE) (Hofmann et al., 1999; Herold et al., 2007). RAGE is a member of the immunoglobulin superfamily and present on numerous cell types. It has been shown to play crucial roles in a variety of pathophysiological situations, such as wound healing, atherosclerotic lesion development, tumor growth and metastasis, systemic amyloidosis, and Alzheimer disease (Bierhaus et al., 2005). RAGE/S100 interaction has been considered a very attractive model to explain how RAGE and its proinflammatory ligand contribute to the pathophysiology of several inflammatory diseases.
Beside the above mentioned receptors a number of other cell surface binding sites specific for S100A8/A9 have been reported, such as novel carboxylated glycans (Srikrishna et al., 2001), heparan sulfate glycosaminoglycans (Robinson et al., 2002), beta2-integrin (Newton and Hogg, 1998), and the fatty acid transporter FAT/CD36 (Kerkhoff et al., 2001). Therefore, the cell surface receptor of S100A8/A9 is still in debate.
Interestingly, the growth-stimulatory activity of S100A8/A9 has been demonstrated to be mediated by binding to the receptor of advanced glycation end products (RAGE) (Ghavami et al., 2008b; Turovskaya et al., 2008; Gebhardt et al., 2008). It is likely to speculate that the selective up-regulation of S100 proteins may be of importance for survival and proliferation of metastasizing cancer cells.
S100A8/A9 complexes that are secreted from phorbolester-stimulated neutrophil-like HL-60 cells have been shown to carry the eicosanoid precursor arachidonic acid (Kerkhoff et al., 1999). The S100A8/A9-arachidonic acid complex is recognized by the fatty acid transporter FAT/CD36, and the fatty acid is rapidly taken up (Kerkhoff et al., 2001). Endothelial cells as well as neutrophils themselves utilize both endogenous and exogenous arachidonic acid for transcellular production of eicosanoids (Sala et al., 1999). Therefore, the secreted S100A8/A9-AA complex may serve as a transport protein to move AA to its target cells. This may represent a mechanism by which AA-derived eicosanoids are synthesized in a cooperative manner between different cell species due to environmental cues.
S100A8/A9 displays apoptosis-inducing activity against various tumor cells (Yui et al., 1995; Yui et al., 2002; Ghavami et al., 2004; Ghavami et al., 2008a; Kerkhoff and Ghavami, 2009; Ghavami et al., 2009; Ghavami et al., 2010). It was speculated that this activity was due to the ability to bind divalent metal ions including Zn²⁺, Mn²⁺ and Cu²⁺ at sites that are distinct from Ca²⁺-binding sites. However, a number of recent reports now indicate that S100A8/A9 exerts its activity by both chelation of trace metal ions such as Zn²⁺ and cell surface receptor mediated pathways.
Although a number of receptors have been shown to bind S100A8/A9, the nature of the receptor involved in S100A8/A9-induced cell death remains to be elucidated. Experiments with certain cell lines either deficient for or over expressing components of the death signaling machinery as well as RAGE gene silencing and blocking RAGE-specific antibody approaches excluded both RAGE and the classical death receptor to be involved in S100A8/A9-induced cell death, even though S100A8/A9 can specifically bind to cancer cells and RAGE mediates the growth-promoting activity obvious at low micromolar concentrations of S100A8/A9. Clearly, investigations to identify the receptor involved in S100A8/A9-induced cell death are critical.

Overall, the S100 proteins share significant sequential homology in the EF-hand motifs, but are least conserved in the hinge region. This region is proposed to provide for specific interaction with target proteins (Groves et al., 1998; Zimmer et al., 2003; Santamaria-Kisiel et al., 2006; Fernandez-Fernandez et al., 2008; van Dieck et al., 2009). The availability of high-resolution S100-target structures has highlighted important structural features that contribute to S100 protein functional specificity (Bhattacharya et al., 2003).
The functional diversification of S100 proteins is achieved by their specific cell- and tissue-expression patterns, structural variations, different metal ion binding properties (Ca²⁺, Zn²⁺ and Cu²⁺) as well as their ability to form homo-, hetero- and oligomeric assemblies (Hunter and Chazin, 1998; Osterloh et al., 1998; Pröpper et al., 1999; Tarabykina et al., 2001; Moroz et al., 2003; Fritz et al., 2010). Although the function of S100 proteins in cancer cells in most cases is still unknown, the specific expression patterns of these proteins are a valuable diagnostic tool.