Structural basis of the Ca(2+)-dependent association between S100C (S100A11) and its target, the N-terminal part of annexin I.

BACKGROUND: S100C (S100A11) is a member of the S100 calcium-binding protein family, the function of which is not yet entirely clear, but may include cytoskeleton assembly and dynamics. S100 proteins consist of two EF-hand calcium-binding motifs, connected by a flexible loop. Like several other members of the family, S100C forms a homodimer. A number of S100 proteins form complexes with annexins, another family of calcium-binding proteins that also bind to phospholipids. Structural studies have been undertaken to understand the basis of these interactions. RESULTS: We have solved the crystal structure of a complex of calcium-loaded S100C with a synthetic peptide that corresponds to the first 14 residues of the annexin I N terminus at 2.3 A resolution. We find a stoichiometry of one peptide per S100C monomer, the entire complex structure consisting of two peptides per S100C dimer. Each peptide, however, interacts with both monomers of the S100C dimer. The two S100C molecules of the dimer are linked by a disulphide bridge. The structure is surprisingly close to that of the p11-annexin II N-terminal peptide complex solved previously. We have performed competition experiments to try to understand the specificity of the S100-annexin interaction. CONCLUSIONS: By solving the structure of a second annexin N terminus-S100 protein complex, we confirmed a novel mode of interaction of S100 proteins with their target peptides; there is a one-to-one stoichiometry, where the dimeric structure of the S100 protein is, nevertheless, essential for complex formation. Our structure can provide a model for a Ca(2+)-regulated annexin I-S100C heterotetramer, possibly involved in crosslinking membrane surfaces or organising membranes during certain fusion events.

Phospholipid vesicle binding and aggregation by four novel fish annexins are differently regulated by Ca2+.

Four members of the annexin family, herein referred to as max (for medaka annexin) 1-4, have recently been identified through hybridization cloning in the killifish Oryzias latipes (D. Osterloh, J. Wittbrodt and V. Gerke, Characterization and developmentally regulated expression of four annexins in the killifish medaka. DNA and Cell Biol., in press). These annexins which are expressed in a developmentally regulated manner are present as a maternal pool in unfertilized eggs of another fish species, Misgurnus fossilis, and it has been proposed that they play a role in the Ca2+-regulated exocytosis of cortical granules occurring after fertilization. To characterize biochemical properties of the medaka proteins possibly relevant to their function in early development, we analyzed the ability of recombinantly expressed max 1-4 to interact with the principal structures of the egg cortex, phospholipid membranes and actin filaments. We show that all medaka annexins bind to acidic phospholipids in a Ca2+-regulated manner, although exhibiting different Ca2+ sensitivities. All medaka annexins, but max 1, are also capable of inducing, in a Ca2+-dependent manner, phospholipid vesicle aggregation, albeit only max 3 displays this activity at Ca2+ concentrations met in stimulated (i.e. fertilized) eggs. Max 3 is also the only medaka annexin able to interact with F-actin in the presence of Ca2+. These data identify by biochemical criteria max 3 as a close relative of the mammalian annexins I and II, thus supporting previous sequence-based comparisons. Max 3 is therefore the prime annexin candidate for being involved in cortical granule exocytosis, possibly by providing granule granule, granule plasma membrane and/or granule cytoskeleton contacts.

Hydrophobic residues in the C-terminal region of S100A1 are essential for target protein binding but not for dimerization.

S100 proteins are a family of small dimeric proteins characterized by two EF hand type Ca2+ binding motifs which are flanked by unique N- and C-terminal regions. Although shown unequivocally in only a few cases S100 proteins are thought to function by binding to, and thereby regulating, cellular target proteins in a Ca2+ dependent manner. To describe for one member of the family, S100A1, structural requirements underlying target protein binding, we generated specifically mutated S100A1 derivatives and characterized their interaction with the alpha subunit of the actin capping protein CapZ shown here to represent a direct binding partner for S100A1. Chemical cross-linking, ligand blotting and fluorescence emission spectroscopy reveal that removal of, or mutations within, the sequence encompassing residues 88-90 in the unique C-terminal region of S100A1 interfere with binding to CapZ alpha and to TRTK-12, a synthetic CapZ alpha peptide. The S100A1 sequence identified contains a cluster of three hydrophobic residues (Phe-88, Phe-89 and Trp-90) at least one of which--as revealed by qualitative phenyl Sepharose binding and hydrophobic fluorescent probe spectroscopy--is exposed on the protein surface of Ca2+ bound S100A1. As homologous hydrophobic residues in the closely related S100B protein were shown by NMR spectroscopy of Ca(2+)-free S100B dimers to provide intersubunit contacts [Kilby P.M., van Eldik L.J., Roberts G.C.K. The solution structure of the bovine S100B dimer in the calcium-free state. Structure 1996; 4: 1041-1052; Drohat A.C., Amburgey J.C., Abildgaard F., Starich M.R., Baldisseri D., Weber D.J. Solution structure of rat apo-S100B (beta beta) as determined by NMR spectroscopy. Biochemistry 1996; 35: 11,577-11,588], we characterized the physical state of the various S100A1 derivatives. Analytical gel filtration and chemical cross-linking show that dimer formation is not compromised in S100A1 mutants lacking residues 88-90 or containing specific amino acid substitutions in this sequence. Thus a cluster of hydrophobic residues in the C-terminal region of S100A1 is essential for target protein binding but dispensable for dimerization, a situation possibly met in other S100 proteins as well.

Characterization and developmentally regulated expression of four annexins in the killifish medaka.

Annexins are Ca2+-regulated membrane binding proteins implicated in a wide range of membrane-related and signal transduction events, including the endocytosis of membrane receptors and Ca2+-regulated as well as constitutive secretion. To date, 10 unique members of this multigene family have been identified in a variety of cell types and tissues of higher vertebrates, with different members showing distinct tissue distributions in the adult organisms. To establish whether annexins also function in embryonic development, we analyzed the expression pattern during vertebrate morphogenesis using the medaka fish Oryzias latipes as a model system. From a larval medaka cDNA library, we isolated four types of clones, which were shown by sequence analysis to encode four different annexins (herein referred to as max 1-4). A comparison with known annexin sequences in the databases revealed that two medaka annexins (max 1 and 2) are highly similar in sequence to mammalian annexins V and IV, respectively, whereas the other two medaka annexins (max 3 and 4) are probably novel members of the family most closely related to mammalian annexins I and XI. Using whole-mount RNA in situ hybridization, we showed that the expression of the different medaka annexins during embryogenesis was strictly regulated at both the spatial and the temporal level. High levels of max 1, 2, and 3 transcripts were present in the developing stomach, gut, liver, air-bladder, and rectum during somitogenesis, thus identifying the digestive tract as the prime region of annexin expression. Interestingly, two structures playing crucial roles in neuronal patterning showed a distinct expression of annexins. The mesendoderm of the anterior prechordal plate of neurula-stage embryos was a site of max 4 transcription, and the floor plate of somitogenesis-stage embryos showed expression of max 2 and 3 to differing rostrocaudal extends along the brain and spinal cord. These results suggest specific functions of different annexins during vertebrate morphogenesis.

Role of the C-terminal extension in the interaction of S100A1 with GFAP, tubulin, the S100A1- and S100B-inhibitory peptide, TRTK-12, and a peptide derived from p53, and the S100A1 inhibitory effect on GFAP polymerization.

Whereas native and recombinant S100A1 inhibited GFAP assembly, a truncated S100A1 lacking the last six C-terminal residues (Phe88-Ser93) (S100A1Delta88-93) proved unable to do so. The inhibitory effects of native and recombinant S100A1 on GFAP assembly were blocked by both TRTK-12, a synthetic peptide derived from the alpha-subunit of the actin capping protein, CapZ, and a synthetic peptide derived from the tumor-suppressor protein, p53, in a dose-dependent manner. By fluorescent spectroscopy, TRTK-12 and the p53 peptide, like GFAP and tubulin, caused a dose- and Ca2+-dependent blue-shift of the fluorescence maximum of acrylodan-S100A1. In contrast, GFAP, tubulin, TRTK-12, or the p53 peptide caused no significant changes in the fluorescence spectrum of acrylodan-S100A1Delta88-93. By chemical crosslinking, both TRTK-12 and the p53 peptide strongly reduced or blocked the formation of GFAP-S100A1 or tubulin-S100A1 complexes, respectively, and S100A1Delta88-93 was unable to complex with tubulin, whereas a remarkably reduced complexation of GFAP with the truncated protein was observed. All the above observations show that the C-terminal extension of S100A1 is an essential part of the S100A1 site implicated in the recognition of GFAP, tubulin, p53, and the alpha-subunit of CapZ.

The crystal structure of a complex of p11 with the annexin II N-terminal peptide.

The aggregation and membrane fusion properties of annexin II are modulated by the association with a regulatory light chain called p11.p11 is a member of the S100 EF-hand protein family, which is unique in having lost its calcium-binding properties. We report the first structure of a complex between p11 and its cognate peptide, the N-terminus of annexin II, as well as that of p11 alone. The basic unit for p11 is a tight, non-covalent dimer. In the complex, each annexin II peptide forms hydrophobic interactions with both p11 monomers, thus providing a structural basis for high affinity interactions between an S100 protein and its target sequence. Finally, p11 forms a disulfide-linked tetramer in both types of crystals thus suggesting a model for an oxidized form of other S100 proteins that have been found in the extracellular milieu.