The S100 calcium-binding protein family of vertebrate, metal-regulated proteins plays pivotal roles in a wide variety of intracellular and extracellular functions including cell growth, inflammation, membrane remodeling and chemotaxis [1, 2] and are implicated in many diseases including cancer [3–6]. Hallmarks of the S100 protein family include their small size (~ 100 aa) and the presence of a canonical (calmodulin-like) EF-hand motif and a non-canonical (S100-specific) EF-hand motif . Additional complexity in S100 proteins is derived from their ability to adopt a non-covalent anti-parallel homo/heterodimers that can further associate to yield higher order multimers such as tetramers and hexamers [3, 5, 8, 9].
While most S100 proteins share moderate sequence identity, S100A7 and S100A15 are 93% identical in sequence yet intriguingly display divergent functions. S100A7, for example, is implicated in inflammatory skin diseases and in several cancers including breast cancer [4, 10–13] where it is associated with aggressive estrogen receptor negative tumors and poor prognosis [14, 15]. Intracellular S100A7 has also been shown to interact with the multifunctional signaling protein Jab1, resulting in translocation of Jab1 to the nucleus where it mediates S100A7 tumorigenic effects [16, 17]. While S100A15 is also highly expressed in psoriatic lesions [18, 19], it displays distinct localizations in skin and breast, divergent functions in epithelial maturation and skin inflammation [19, 20] and it is currently unknown whether S100A15 contributes to breast cancer progression . S100A7 and S100A15 are also capable of binding different surface receptors . S100A7 interacts with RAGE (receptor for advanced glycated end products) to mediate chemotaxis of leukocytes [20, 22] while S100A15 appears to mediate chemotaxis through an unidentified G protein coupled receptor .
The ability of many S100 proteins to coordinate zinc is central to their function . For example, the ability of S100A7 to engage RAGE and exert antimicrobial effects are both dependent on zinc . These observations suggest a regulatory role although the zinc coordination geometry and affinity suggest a structural role . It is not yet known whether zinc binding plays a structural or regulatory role in S100A7 and S100A15 function. To our knowledge, there is no data to suggest S100A15 function is zinc-dependent. Thus far, two types of S100-family zinc sites have been proposed . In the first type (‘His-Zn’), the zinc is coordinated by 3 histidines and an aspartate (i.e. S100A7 and S100A12) or with four histidines (i.e. S100B). The second type (‘Cys-Zn’) is proposed to contain primarily cysteines as the zinc-coordinating residues though no structure of an S100 ‘Cys-Zn’ binding site has been reported. Ultimately defining the zinc binding capacity for S100A15 is crucial to understanding its function and appropriately categorizing it within the S100 family of proteins. To this end, we report the structural characterization of S100A15 and an Asp24Gly variant of S100A7 to 1.7 Å and 1.6 Å, respectively. These data reveal an unexpected structural compensation to retain zinc binding in S100A15 and a structural rationale for the inability of S100A15 to coordinate RAGE and the prediction that S100A15 will not bind Jab1.