Thomas H. Charpentier

Structure of Ca2+-Bound S100A4 and Its Interaction with Peptides Derived from Nonmuscle Myosin-IIA†

S100A4, also known as mts1, is a member of the S100 family of Ca2+-binding proteins that is directly involved in tumor invasion and metastasis via interactions with specific protein targets, including nonmuscle myosin-IIA (MIIA). Human S100A4 binds two Ca2+ ions with the typical EF-hand exhibiting an affinity that is nearly 1 order of magnitude tighter than that of the pseudo-EF-hand. To examine how Ca2+ modifies the overall organization and structure of the protein, we determined the 1.7 A crystal structure of the human Ca2+-S100A4. Ca2+ binding induces a large reorientation of helix 3 in the typical EF-hand. This reorganization exposes a hydrophobic cleft that is comprised of residues from the hinge region,helix 3, and helix 4, which afford specific target recognition and binding. The Ca2+-dependent conformational change is required for S100A4 to bind peptide sequences derived from the C-terminal portion of the MIIA rod with submicromolar affinity. In addition, the level of binding of Ca2+ to both EF-hands increases by 1 order of magnitude in the presence of MIIA. NMR spectroscopy studies demonstrate that following titration with a MIIA peptide, the largest chemical shift perturbations and exchange broadening effects occur for residues in the hydrophobic pocket of Ca2+-S100A4. Most of these residues are not exposed in apo-S100A4 and explain the Ca2+ dependence of formation of theS100A4-MIIA complex. These studies provide the foundation for understanding S100A4 target recognition and may support the development of reagents that interfere with S100A4 function.

Divalent metal ion complexes of S100B in the absence and presence of pentamidine.

As part of an effort to inhibit S100B, structures of pentamidine (Pnt) bound to Ca(2+)-loaded and Zn(2+),Ca(2+)-loaded S100B were determined by X-ray crystallography at 2.15 A (R(free)=0.266) and 1.85 A (R(free)=0.243) resolution, respectively. These data were compared to X-ray structures solved in the absence of Pnt, including Ca(2+)-loaded S100B and Zn(2+),Ca(2+)-loaded S100B determined here (1.88 A; R(free)=0.267). In the presence and absence of Zn(2+), electron density corresponding to two Pnt molecules per S100B subunit was mapped for both drug-bound structures. One Pnt binding site (site 1) was adjacent to a p53 peptide binding site on S100B (+/-Zn(2+)), and the second Pnt molecule was mapped to the dimer interface (site 2; +/-Zn(2+)) and in a pocket near residues that define the Zn(2+) binding site on S100B. In addition, a conformational change in S100B was observed upon the addition of Zn(2+) to Ca(2+)-S100B, which changed the conformation and orientation of Pnt bound to sites 1 and 2 of Pnt-Zn(2+),Ca(2+)-S100B when compared to Pnt-Ca(2+)-S100B. That Pnt can adapt to this Zn(2+)-dependent conformational change was unexpected and provides a new mode for S100B inhibition by this drug. These data will be useful for developing novel inhibitors of both Ca(2+)- and Ca(2+),Zn(2+)-bound S100B.

The effects of CapZ peptide (TRTK-12) binding to S100B-Ca2+ as examined by NMR and X-ray crystallography.

Structure-based drug design is underway to inhibit the S100B-p53 interaction as a strategy for treating malignant melanoma. X-ray crystallography was used here to characterize an interaction between Ca(2)(+)-S100B and TRTK-12, a target that binds to the p53-binding site on S100B. The structures of Ca(2+)-S100B (1.5-A resolution) and S100B-Ca(2)(+)-TRTK-12 (2.0-A resolution) determined here indicate that the S100B-Ca(2+)-TRTK-12 complex is dominated by an interaction between Trp7 of TRTK-12 and a hydrophobic binding pocket exposed on Ca(2+)-S100B involving residues in helices 2 and 3 and loop 2. As with an S100B-Ca(2)(+)-p53 peptide complex, TRTK-12 binding to Ca(2+)-S100B was found to increase the proteins Ca(2)(+)-binding affinity. One explanation for this effect was that peptide binding introduced a structural change that increased the number of Ca(2+) ligands and/or improved the Ca(2+) coordination geometry of S100B. This possibility was ruled out when the structures of S100B-Ca(2+)-TRTK-12 and S100B-Ca(2+) were compared and calcium ion coordination by the protein was found to be nearly identical in both EF-hand calcium-binding domains (RMSD=0.19). On the other hand, B-factors for residues in EF2 of Ca(2+)-S100B were found to be significantly lowered with TRTK-12 bound. This result is consistent with NMR (15)N relaxation studies that showed that TRTK-12 binding eliminated dynamic properties observed in Ca(2+)-S100B. Such a loss of protein motion may also provide an explanation for how calcium-ion-binding affinity is increased upon binding a target. Lastly, it follows that any small-molecule inhibitor bound to Ca(2+)-S100B would also have to cause an increase in calcium-ion-binding affinity to be effective therapeutically inside a cell, so these data need to be considered in future drug design studies involving S100B.

Design of Inhibitors for S100B

S100B interacts with the p53 protein in a calcium-dependent manner and down-regulates its function as a tumor suppressor. Therefore, inhibiting the S100B-p53 interaction represents a new approach for restoring functional wild-type p53 in cancers with elevated S100B such as found in malignant melanoma. A discussion of the biological rational for targeting S100B and a description of methodologies relevant to the discovery of compounds that inhibit S100B-p53 binding, including computational techniques, structural biology techniques, and cellular assays, is presented.

Small molecules bound to unique sites in the target protein binding cleft of calcium-bound S100B as characterized by nuclear magnetic resonance and X-ray crystallography.

Structural studies are part of a rational drug design program aimed at inhibiting the S100B-p53 interaction and restoring wild-type p53 function in malignant melanoma. To this end, structures of three compounds (SBi132, SBi1279, and SBi523) bound to Ca(2+)-S100B were determined by X-ray crystallography at 2.10 A (R(free) = 0.257), 1.98 A (R(free) = 0.281), and 1.90 A (R(free) = 0.228) resolution, respectively. Upon comparison, SBi132, SBi279, and SBi523 were found to bind in distinct locations and orientations within the hydrophobic target binding pocket of Ca(2+)-S100B with minimal structural changes observed for the protein upon complex formation with each compound. Specifically, SBi132 binds nearby residues in loop 2 (His-42, Phe-43, and Leu-44) and helix 4 (Phe-76, Met-79, Ile-80, Ala-83, Cys-84, Phe-87, and Phe-88), whereas SBi523 interacts with a separate site defined by residues within loop 2 (Ser-41, His-42, Phe-43, Leu-44, Glu-45, and Glu-46) and one residue on helix 4 (Phe-87). The SBi279 binding site on Ca(2+)-S100B overlaps the SBi132 and SBi523 sites and contacts residues in both loop 2 (Ser-41, His-42, Phe-43, Leu-44, and Glu-45) and helix 4 (Ile-80, Ala-83, Cys-84, Phe-87, and Phe-88). NMR data, including saturation transfer difference (STD) and (15)N backbone and (13)C side chain chemical shift perturbations, were consistent with the X-ray crystal structures and demonstrated the relevance of all three small molecule-S100B complexes in solution. The discovery that SBi132, SBi279, and SBi523 bind to proximal sites on Ca(2+)-S100B could be useful for the development of a new class of molecule(s) that interacts with one or more of these binding sites simultaneously, thereby yielding novel tight binding inhibitors specific for blocking protein-protein interactions involving S100B.

In vitro screening and structural characterization of inhibitors of the S100B-p53 interaction

S100B is highly over-expressed in many cancers, including malignant melanoma. In such cancers, S100B binds wild-type p53 in a calcium-dependent manner, sequestering it, and promoting its degradation, resulting in the loss of p53-dependent tumor suppression activities. Therefore, S100B inhibitors may be able to restore wild-type p53 levels in certain cancers and provide a useful therapeutic strategy. In this regard, an automated and sensitive fluorescence polarization competition assay (FPCA) was developed and optimized to screen rapidly for lead compounds that bind Ca(2+)-loaded S100B and inhibit S100B target complex formation. A screen of 2000 compounds led to the identification of 26 putative S100B low molecular weight inhibitors. The binding of these small molecules to S100B was confirmed by nuclear magnetic resonance spectroscopy, and additional structural information was provided by x-ray crystal structures of several compounds in complexes with S100B. Notably, many of the identified inhibitors function by chemically modifying Cys84 in protein. These results validate the use of high-throughput FPCA to facilitate the identification of compounds that inhibit S100B. These lead compounds will be the subject of future optimization studies with the ultimate goal of developing a drug with therapeutic activity for the treatment of malignant melanoma and/or other cancers with elevated S100B.

Getting a grip on calcium regulation

Temporally and spatially regulated Ca2+ signals control numerous physiological processes. In most cells in higher animals, the plasma membrane (PM) Na/Ca exchanger (NCX) helps manage the cytosolic Ca2+ concentration ([Ca2+]CYT) and Ca2+ stored in the sarco-/endoplasmic reticulum (S/ER), thereby influencing Ca2+ signaling (1–3). This tightly regulated transporter can move one Ca2+ ion either out of or into cells in exchange for three Na+. The direction of net Ca2+ movement depends on the prevailing membrane potential (V M) as well as the Na+ and Ca2+ concentration gradients because there is net charge transfer. Cardiac myocytes have large dynamic swings in V M and [Ca2+]CYT during each heartbeat, and NCX plays an especially interesting role: It can mediate Ca2+ entry during the upstroke of the action potential, help maintain the elevated [Ca2+]CYT and contraction during the action potential plateau, and then extrude Ca2+ during repolarization and diastole (1). It is noteworthy that NCX is regulated by cytosolic Na+ and Ca2+ at sites that do not directly participate in the ion translocation (4, 5). A rise in cytosolic Na+ rapidly stimulates and then inactivates the exchanger (6); in contrast, cytosolic Ca2+ activates the exchanger and relieves the Na+-dependent inactivation (4, 5). Two elegant studies by Philipson, Abramson, and colleagues, including one in this issue of PNAS (7), provide novel insight into how Ca2+ binds to and alters the conformation of the Ca2+ regulatory sites in the cardiac/neuronal NCX (NCX type-1) (7, 8).

Hydrocarbon-Stapled Helices: A Novel Approach for Blocking Protein-Protein Interactions

Modulating protein–protein interfaces (PPIs) for therapeutic intervention has long been a vision of the research and pharmaceutical community, and approaches taken for inhibiting the Hdm2– p53 interaction illustrates the possibilities and problems associated with targeting PPIs. Normally, the tumor suppressor protein p53 functions as a molecular sentinel, which upon DNA damage, regulates genes that control cell-cycle arrest, apoptosis, senescence, differentiation, and DNA repair. The elimination or mutation of the p53 protein causes increased tumor formation, which is observed in p53 homozygous knockout mice and in Li-Fraumeni disease where germline mutations in p53 increases patients predisposition for cancer. Understandably, p53 is itself tightly regulated to allow normal cell growth and differentiation. To this end, p53 induces the expression of its own negative regulators including the ubiquitin E3 ligase, Hdm2, which binds p53 directly, inhibiting its intrinsic activities, and promoting its ubiquitin-dependent degradation by the proteosome. However, if Hdm2 is overexpressed, as is found in 7% of all cancers, then excessive p53 degradation occurs and its tumor suppression function is lost. While p53 is deleted or mutated in ~50% of all tumors, there are many tumors that have wild-type p53. In several wild-type p53 cancers, the inactivation of p53 occurs because of the overexpression or the aberrant regulation of Hdm2, which is thought to contribute significantly to the disease. Therefore, inhibiting the Hdm2–p53 protein–protein interaction is now a widely accepted therapeutic strategy to restore p53 protein levels and correspondingly its function as a tumor suppressor in such cancers. This therapeutic approach was first validated in cellular assays by microinjection of antibodies that blocked the Hdm2–p53 interaction in mutant Ras transformed rat thyroid ephithelial cells and by the use of anti-Hdm2 siRNA to knockdown Hdm2 in breast carcinoma cells (MCF7) and osteosarcoma cells (JAR). In such validation experiments, p53 protein levels increased, and p53-dependent apoptosis activity was restored as predicted. As a result, numerous groups began making synthetic inhibitors of the Hdm2–p53 interaction, including peptidomimetics and small-molecule inhibitors with the goal of engineering a new cancer drug. Unfortunately, there were varying degrees of success in restoring functional wild-type p53 protein in tumor cells with small molecule inhibitors. As with other PPIs, it was thought that specifically targeting the Hdm2–p53 interaction with small molecules was difficult because of the large binding surface area at the protein–protein interface with multiple contacts involved. Whereas peptidomimetics have overcome some of these issues, others arise such as their instability, low affinity, and lack of cell permeability in many cases. However, the Verdine Lab at the Harvard Medical School has provided an elegant and novel peptidomimetic strategy for inhibiting the Hdm2–p53 and other protein–protein interactions, which has overcome many of the problems associated with the previously available small molecule and peptidomimetic approaches. In a recent paper published in Journal of American Chemical Society, the Verdine group developed and implemented a new technique termed “peptide stapling” and have applied it to inhibiting the Hdm2–p53 interaction. Structural data revealed that the p53 peptide forms an amphipathic a-helix upon binding a hydrophobic cleft in Hdm2, with three residues on the same face of the a-helical peptide (F19, W23, and L26) interacting directly with Hdm2. Taking advantage of the fact that the region of p53 bound to Hdm2 is helical, the Verdine group strategically incorporates non-natural a,a-disubstituted amino acids containing olefinic side chains into the peptide and then cross-links the alkyl side chains using ruthenium-catalyzed ring-closing olefin metathesis. This reaction in effect “staples” the p53 peptide, which would otherwise adopt numerous conformations in solution, into one that nicely adopts a single a-helical secondary structure (Figure 1). In such a design, it is critically important that the modification does not occlude binding, as judged by examining 3D structures of the peptide–protein complex a priori. Thus, the affect of the hydrocarbon staple was examined by synthesizing a series of modified p53 peptides with varied positioning of the stereospecific modified amino acid. In addition, R or S stereochemistry at either one (i, i+4) or two (i, i+7) turns of the a-helix were used as were varied lengths of the linkers. After such molecules were synthesized, the various peptides were crosslinked (that is, stapled) and tested for [a] Dr. P. T. Wilder, T. H. Charpentier, Dr. D. J. Weber Department of Biochemistry & Molecular Biology 108 North Greene Street Baltimore, MD 21201 Fax: (+1)410-706-0458 E-mail : dweber@umaryland.edu