Molecular recognition of protein surfaces: high affinity ligands for the CBP KIX domain

Potent and specific inhibitors of protein.protein interactions have potential both as therapeutic compounds and biological tools, yet discovery of such molecules remains a challenge. Our laboratory has recently described a strategy, called protein grafting, for the identification of miniature proteins that bind protein surfaces with high affinity and specificity and inhibit the formation of protein.protein complexes. In protein grafting, those residues that comprise a functional alpha-helical binding epitope are stabilized on the solvent-exposed alpha-helical face of the small yet stable protein avian pancreatic polypeptide (aPP). Here we use protein grafting in combination with molecular evolution by phage display to identify phosphorylated peptide ligands that recognize the shallow surface of CBP KIX with high nanomolar to low micromolar affinity. Furthermore, we show that grafting of the CBP KIX-binding epitope of CREB KID onto the aPP scaffold yields molecules capable of high affinity recognition of CBP KIX even in the absence of phosphorylation. Importantly, both classes of designed ligands exhibit high specificity for the target CBP KIX domain over carbonic anhydrase and calmodulin, two unrelated proteins that bind hydrophobic or alpha-helical molecules that might be encountered in vivo.     

Paralog-selective ligands for bcl-2 proteins

There is considerable current interest in molecules that bind intra- or extracellular protein surfaces and inhibit protein-protein interactions. Previously we have reported that miniature proteins based on pancreatic-fold polypeptides can recognize even shallow alpha-helix binding clefts with high affinity and selectivity against unrelated proteins. One such miniature protein, PPBH3-1, binds the anti-apoptotic protein paralogs Bcl-2 and Bcl-XL with nanomolar affinity and a DeltaDeltaG = 1.2 kcal.mol-1 preference for Bcl-XL. Here we describe the directed evolution of PPBH3-1 into two new miniature proteins, PPBH3-5 and PPBH3-6, whose paralog specificity is reversed relative to PPBH3-1. PPBH3-5 and PPBH3-6 bind Bcl-2 with nanomolar affinity and a DeltaDeltaG = 0.9-1.3 kcal.mol-1 preference for Bcl-2 over Bcl-XL. Experiments with Bcl-XL variants suggest that PPBH3-5 and PPBH3-6 achieve high paralog specificity by exploiting subtle structural or electrostatic differences in the Bcl-2 and Bcl-XL molecular landscapes. PPBH3-5 and PPBH3-6 may have unique applications as early examples of nonnatural ligands that interact selectively with Bcl-2 proteins.     

Miniature homeodomains: high specificity without an N-terminal arm

Recently, we described a strategy for the design of miniature proteins that bind DNA and protein surfaces with high affinity and selectivity. This strategy involves identifying the functional epitope required for macromolecular recognition by a natural protein and presenting it on a small, stable protein scaffold. In previous work, high-affinity DNA recognition was achieved only when the miniature protein contained the complete functional epitope. Here we report a miniature homeodomain that recognizes its 6-bp target site in the nanomolar concentration range at 25 degrees C, despite the absence of DNA contact residues located along the homeodomain N-terminal arm. We conclude that miniature proteins can achieve high affinity and selectivity for DNA by design even when the functional epitope is incomplete by using pre-organization to effectively compensate for lost protein-DNA contacts. In this case it has been possible to miniaturize both the recognition surface and the structural framework of a globular protein fold.     

Encodable activators of SRC family kinases

There is considerable current interest in the design of encodable molecules that regulate intracellular protein circuitry and/or activity, ideally with a high level of specificity. Src homology 3 (SH3) domains are ubiquitous components of multidomain signaling proteins, including many kinases, and are attractive drug targets because of the important role their interactions play in diseases as diverse as cancer, osteoporosis, and inflammation. Here we describe a set of miniature proteins that recognize distinct SH3 domains from Src family kinases with high affinity. Three of these molecules discriminate effectively between the SH3 domains of Src and Fyn, which are expressed ubiquitously, and two of these three activate Hck kinase with potencies that rival HIV Nef, one of the most potent kinase activators known. These results suggest that miniature proteins represent a viable, encodable strategy for selective activation of Src family kinases in a variety of cell types.     

Rewiring kinase specificity with a synthetic adaptor protein

Signaling cascades are managed in time and space by interactions between and among proteins. These interactions are often aided by adaptor proteins, which guide enzyme-substrate pairs into proximity. Miniature proteins are a class of small, well-folded protein domains possessing engineered binding properties. Here we made use of two miniature proteins with complementary binding properties to create a synthetic adaptor protein that effectively redirects a ubiquitous signaling event: tyrosine phosphorylation. We report that miniature-protein-based adaptor 3 uses templated catalysis to redirect the Src family kinase Hck to phosphorylate hDM2, a negative regulator of the p53 tumor suppressor and a poor Hck substrate. Phosphorylation occurs with multiple turnover and at a single site targeted by c-Abl kinase in the cell.     

Increasing the kinase specificity of k252a by protein surface recognition

[reaction: see text] Here we describe a miniature protein (1) that presents the cAMP-dependent protein kinase (PKA) recognition epitope found within the heat-stable Protein Kinase Inhibitor protein (PKI) and a miniature protein conjugate (1-K252a) in which 1 is joined covalently to the high-affinity but nonselective kinase inhibitor K252a. Miniature protein 1 recognizes PKA with an affinity that rivals that of PKI and, in the context of 1-K252a, leads to a dramatic increase in kinase specificity.     

De novo design of a stable N-terminal helical foldamer

A peptide NTH-18 was synthesized in which a N-terminal helix is stabilised by two crossed disulfide bonds to a C-terminal extension. The design was inspired by the structure of the neurotoxic peptide apamin, which has previously been used to stabilise helices in miniature enzymes. CD- and NMR-spectroscopy indicated that NTH-18 adopted a fold similar to that found in apamin. However, the arrangement of the elements of secondary structures was inverted relative to apamin; a N-terminal alpha-helix was connected by a reverse turn to a C-terminal extension of non-canonical secondary structure. NTH-18 displayed significant stability to heat and changes of pH. The high definition of the N-terminal end of the alpha-helix of NTH-18 should make this peptide a useful vehicle to stabilise alpha-helices in proteins with applications in protein engineering and molecular recognition.     

Inhibition of beta-amyloid fibrillization by directed evolution of a beta-sheet presenting miniature protein

We describe the directed evolution of a miniature beta-sheet protein for targeting beta-amyloid oligomers implicated in Alzheimer's disease. Circular dichroism spectroscopy, thermal denaturation experiments, and immunoglobulin binding assays established that our beta-amyloid-targeted miniature protein, TJ10, presents a well-folded thermostable beta-sheet. TJ10 was found to prevent beta-amyloid fibrillization at stoichiometric concentrations and was also an effective inhibitor at substoichiometric concentrations. Thus our results provide a new and potent beta-sheet chemical template for effectively targeting beta-amyloid while also demonstrating a general strategy for targeting proteins implicated in other amyloidogenic diseases.     

Grafting miniature DNA binding proteins

Miniature proteins serve as leads for biological and medicinal applications by positioning all amino acids necessary for biomolecular recognition on a compact protein structure. Protein grafting was recently used to create miniature helical proteins with high DNA binding affinity and specificity.     

Novel strategy for the design of a new zinc finger: creation of a zinc finger for the AT-rich sequence by alpha-helix substitution

In this communication, a novel strategy for the design of a zinc finger peptide on the basis of alpha-helix substitution has been demonstrated. Sp1HM is a helix-substituted mutant for the wild-type Sp1(zf123) and its alpha-helix of each finger is replaced by that of fingers 4-6 of CF2-II. The circular dichroism spectrum of Sp1HM suggests that Sp1HM has an ordered secondary structure similar to that of Sp1(zf123). From the analyses of the DNA binding affinity and specificity by gel mobility shift assay, it is clearly indicated that Sp1HM specifically binds to the AT-rich sequence (5'-GTA TAT ATA-3') with 3.2 nM dissociation constants. Moreover, the zinc finger peptides for the sequence alternating between the AT- and GC-rich subsites can also be created by the alpha-helix substitution. This strategy is evidently effective and is also more convenient than the phage display method. Consequently, our design method is widely applicable to creating zinc finger peptides with novel binding specificities.     

Development of a potent Bcl-x(L) antagonist based on alpha-helix mimicry

The rational design of low-molecular weight ligands that disrupt protein-protein interactions is still a challenging goal in medicinal chemistry. Our approach to this problem involves the design of molecular scaffolds that mimic the surface functionality projected along one face of an alpha-helix. Using a terphenyl scaffold, which in a staggered conformation closely reproduces the projection of functionality on the surface of an alpha-helix, we designed mimics of the pro-apoptotic alpha-helical Bak-peptide as inhibitors of the Bak/Bcl-xL interaction. This led to the development of a potent Bcl-xL antagonist (KD = 114 nM), whose binding affinity for Bcl-xL was assessed by a fluorescence polarization assay. To determine the binding site of the developed inhibitor we used docking studies and an HSQC-NMR experiment with 15N-labeled Bcl-xL protein. These studies suggest that the inhibitor is binding in the same hydrophobic cleft as the Bak- and Bad-peptides.     

Synthetic non-peptide mimetics of alpha-helices

Proteins in nature fold into native conformations in which combinations of peripherally projected aliphatic, aromatic and ionic functionalities direct a wide range of properties. Alpha-helices, one of the most common protein secondary structures, serve as important recognition regions on protein surfaces for numerous protein-protein, protein-DNA and protein-RNA interactions. These interactions are characterized by conserved structural features within the alpha-helical domain. Rational design of structural mimetics of these domains with synthetic small molecules has proven an effective means to modulate such protein functions. In this tutorial review we discuss strategies that utilize synthetic small-molecule antagonists to selectively target essential protein-protein interactions involved in certain diseases. We also evaluate some of the protein-protein interactions that have been or are potential targets for alpha-helix mimetics.     

Synthetic metal-binding protein surface domains for metal ion-dependent interaction chromatography. II. Immobilization of synthetic metal-binding peptides from metal ion transport proteins as model bioactive protein surface domains.

This preliminary investigation tests the premise that biologically relevant (1) peptide-metal ion interactions, and (2) metal ion-dependent macromolecular recognition events (e.g., peptide-peptide interactions) may be modeled by biomimetic affinity chromatography. Divinylsulfone-activated agarose (6%) was used to immobilize three different synthetic peptides representing metal-binding protein surface domains from the human plasma metal transport protein histidine-rich glycoprotein (HRG). The synthetic peptides represented 1-3 multiple repeat units of the 5-residue sequence (Gly-His-His-Pro-His) found in the C-terminal of HRG. By frontal analyses, immobilized HRG peptides of the type (GHHPH)nG, where n = 1-3, were each found to have a similar binding capacity for both Cu(II) ions and Zn(II) ions (31-38 mumol/ml gel). The metal ion-dependent interaction of a variety of model peptides with each of the immobilized HRG peptide affinity columns demonstrated differences in selectivity despite the similar internal sequence homology and metal ion binding capacity. The immobilized 11-residue HRG peptide was loaded with Cu(II) ions and used to demonstrate selective adsorption and isolation of proteins from human plasma. These results suggest that immobilized metal-binding peptides selected from known solvent-exposed protein surface metal-binding domains may be useful model systems to evaluate the specificity of biologically relevant metal ion-dependent interaction and transfer events in vitro.     

A comprehensive characterization of the T-cell antigen receptor complex composition by microcapillary liquid chromatography-tandem mass spectrometry

It has become apparent that many intracellular signaling processes involve the dynamic reorganization of cellular proteins into complex signaling assemblies that have a specific subunit composition, function, and subcellular location. Since the elements of such assemblies interact physically, multiprotein signaling complexes can be isolated and analyzed. Recent technical advances in highly sensitive protein identification by electrospray-tandem mass spectrometry have dramatically increased the sensitivity with which such analyses can be performed. The T-cell antigen receptor (TCR) is an oligomeric transmembrane protein complex that is essential to T-cell recognition and function. The extracellular protein domains are responsible for ligand binding while intracellular domains generate and transduce signals in response to specific receptor-ligand interactions. We used microbore capillary chromatography-tandem mass spectrometry to investigate the composition of the TCR protein complex isolated from resting and activated cells of the murine T-cell line CD11.3. We identified all the previously known subunits of the TCR/CD3 complex as well as proteins previously not known to associate with the TCR. The catalytic activities of some of these proteins could potentially be used to interfere pharmacologically with TCR signaling.     

Scaffold-independent analysis of RNA-protein interactions: the Nova-1 KH3-RNA complex

We describe a method to analyze the sequence specificity of an RNA-binding domain. The method, which we have named scaffold-independent analysis, reports on the specificity for each nucleotide position within an RNA target, uncoupled from the surrounding structural and sequence context. We expect this information to improve our understanding of protein-RNA interfaces in ssRNA binding domains (e.g., KH or RRM domains) and to be useful to the design of novel protein-RNA recognition surfaces. Our NMR binding assays using the third KH domain of the Nova-1 protein provide a proof-of-principle for the method and novel information on the specificity of this domain for its RNA targets.     

Artificial immunoglobulin G-binding protein mimetic to staphylococcal protein A. Its production and application to affinity purification of immunoglobulin G.

Staphylococcal protein A consists of a single polypeptide with five immunoglobulin G (IgG)-binding domains, which are linked as E-D-A-B-C in this order from the amino terminal. The DNA coding domains A-B were polymerized one to six times linearly, taking advantage of the non-palindromic nucleotide sequence of the AccI recognition site and the resultant DNAs were inserted in pTRP vector carrying trp promoter. The artificial IgG-binding proteins [pA(AB)1-6], which had been expressed in Escherichia coli JM109, were purified by methods involving IgG-Sepharose affinity chromatography. Among pA(AB)1-6 immobilized on cyanogen bromide-Sepharose, pA(AB)4-Sepharose was the highest in IgG-binding capacity at the same level of mg protein per ml gel, about 30% higher than protein A-Sepharose. At 8 mg protein per ml gel, it bound and eluted about 24 mg of IgG from rabbit serum. Its IgG-binding capacities were the highest with porcine, rabbit, human and guinea pig sera, intermediate with bovine, horse and sheep sera and the lowest with mouse, goat, rat and chicken sera.     

Structural basis for chitin recognition by defense proteins: GlcNAc residues are bound in a multivalent fashion by extended binding sites in hevein domains

BACKGROUND: Many plants respond to pathogenic attack by producing defense proteins that are capable of reversible binding to chitin, a polysaccharide present in the cell wall of fungi and the exoskeleton of insects. Most of these chitin-binding proteins include a common structural motif of 30 to 43 residues organized around a conserved four-disulfide core, known as the 'hevein domain' or 'chitin-binding' motif. Although a number of structural and thermodynamic studies on hevein-type domains have been reported, these studies do not clarify how chitin recognition is achieved. RESULTS: The specific interaction of hevein with several (GlcNAc)(n) oligomers has been studied using nuclear magnetic resonance (NMR), analytical ultracentrifugation and isothermal titration microcalorimetry (ITC). The data demonstrate that hevein binds (GlcNAc)(2-4) in 1:1 stoichiometry with millimolar affinity. In contrast, for (GlcNAc)(5), a significant increase in binding affinity is observed. Analytical ultracentrifugation studies on the hevein-(GlcNAc)(5,8) interaction allowed detection of protein-carbohydrate complexes with a ratio of 2:1 in solution. NMR structural studies on the hevein-(GlcNAc)(5) complex showed the existence of an extended binding site with at least five GlcNAc units directly involved in protein-sugar contacts. CONCLUSIONS: The first detailed structural model for the hevein-chitin complex is presented on the basis of the analysis of NMR data. The resulting model, in combination with ITC and analytical ultracentrifugation data, conclusively shows that recognition of chitin by hevein domains is a dynamic process, which is not exclusively restricted to the binding of the nonreducing end of the polymer as previously thought. This allows chitin to bind with high affinity to a variable number of protein molecules, depending on the polysaccharide chain length. The biological process is multivalent.     

Expanding the DNA-recognition repertoire for zinc finger proteins beyond 20 amino acids

The design of DNA binding domains based on the Cys2His2 zinc finger motif has proven to be a successful strategy for the specific recognition of novel DNA sequences. Although considerable effort has been devoted to the generation of zinc finger proteins with widely varying DNA-binding preferences, only a limited number of potential DNA binding sites have been targeted with a high degree of specificity. These restrictions on zinc finger design appear to be a consequence of the limited repertoire of side-chain lengths and functionalities available with the 20 proteinogenic amino acids. To demonstrate that these limitations can be overcome through the use of "unnatural" amino acids, expressed protein ligation was employed to incorporate the amino acid citrulline into a single position within a three-zinc finger protein. As anticipated, the resulting semisynthetic protein specifically recognizes adenine in the appropriate position of its binding site.     

Directed evolution of ATP binding proteins from a zinc finger domain by using mRNA display

Antibodies have traditionally been used for isolating affinity reagents to new molecular targets, but alternative protein scaffolds are increasingly being used for the directed evolution of proteins with novel molecular recognition properties. We have designed a combinatorial library based on the DNA binding domain of the human retinoid-X-receptor (hRXRalpha). We chose this domain because of its small size, stable fold, and two closely juxtaposed recognition loops. We replaced the two loops with segments of random amino acids, and used mRNA display to isolate variants that specifically recognize adenosine triphosphate (ATP), demonstrating a significant alteration of the function of this protein domain from DNA binding to ATP recognition. Many novel independent sequences were recovered with moderate affinity and high specificity for ATP, validating this scaffold for the generation of functional molecules.