Protein-binding Peptides Research Articles

Explainable Machine Learning Model to Accurately Predict Protein-Binding Peptides

Enzymes play key roles in the biological functions of living organisms, which serve as catalysts to and regulate biochemical reaction pathways. Recent studies suggest that peptides are promising molecules for modulating enzyme function due to their advantages in large chemical diversity and well-established methods for library synthesis. Experimental approaches to identify protein-binding peptides are time-consuming and costly. Hence, there is a demand to develop a fast and accurate computational approach to tackle this problem. Another challenge in developing a computational approach is the lack of a large and reliable dataset. In this study, we develop a new machine learning approach called PepBind-SVM to predict protein-binding peptides. To build this model, we extract different sequential and physicochemical features from peptides and use a Support Vector Machine (SVM) as the classification technique. We train this model on the dataset that we also introduce in this study. PepBind-SVM achieves 92.1% prediction accuracy, outperforming other classifiers at predicting protein-binding peptides.

C-Terminal Decarboxylation of Proline-Derived Building Blocks for Protein-Binding Peptides.

Using a set of conformationally restricted Proline-derived Modules (ProMs), our group has recently succeeded in developing inhibitors for the enabled/vasodilator-stimulated phosphoprotein homology 1 (EVH1) domain, which is a key mediator of cell migration and plays an important role in tumor metastasis. While these (formally) pentapeptidic compounds show nanomolecular binding affinities towards EVH1, their drug-like properties and cell permeability need to be further optimized before they can be clinically tested as therapeutic agents against metastasis. In this study, we sought to improve these properties by removing the C-terminal carboxylic acid function of our peptoids, either by late-stage decarboxylation or by direct synthesis. For late-stage decarboxylation of ProM-like systems, a method for reductive halo decarboxylation was optimized and applied to several proline-derived substrates. In this way, a series of new decarboxy ProMs suitable as building blocks for decarboxy EVH1 inhibitors were obtained. In addition, we incorporated decarboxy-ProM-1 into the pentapeptide-like compound Ac[2ClF][ProM-2][Decarb-ProM-1], which showed similar affinity towards EVH1 as the methyl ester derivative (Ac[2Cl-F][ProM-2][ProM1]OMe). However, despite better calculated drug-like properties, this compound did not inhibit chemotaxis in a cellular assay.

Structure-aware machine learning strategies for antimicrobial peptide discovery

Machine learning models are revolutionizing our approaches to discovering and designing bioactive peptides. These models often need protein structure awareness, as they heavily rely on sequential data. The models excel at identifying sequences of a particular biological nature or activity, but they frequently fail to comprehend their intricate mechanism(s) of action. To solve two problems at once, we studied the mechanisms of action and structural landscape of antimicrobial peptides as (i) membrane-disrupting peptides, (ii) membrane-penetrating peptides, and (iii) protein-binding peptides. By analyzing critical features such as dipeptides and physicochemical descriptors, we developed models with high accuracy (86–88%) in predicting these categories. However, our initial models (1.0 and 2.0) exhibited a bias towards α-helical and coiled structures, influencing predictions. To address this structural bias, we implemented subset selection and data reduction strategies. The former gave three structure-specific models for peptides likely to fold into α-helices (models 1.1 and 2.1), coils (1.3 and 2.3), or mixed structures (1.4 and 2.4). The latter depleted over-represented structures, leading to structure-agnostic predictors 1.5 and 2.5. Additionally, our research highlights the sensitivity of important features to different structure classes across models.

Design of protein-binding peptides with controlled binding affinity: the case of SARS-CoV-2 receptor binding domain and angiotensin-converting enzyme 2 derived peptides.

The development of methods able to modulate the binding affinity between proteins and peptides is of paramount biotechnological interest in view of a vast range of applications that imply designed polypeptides capable to impair or favour Protein-Protein Interactions. Here, we applied a peptide design algorithm based on shape complementarity optimization and electrostatic compatibility and provided the first experimental in vitro proof of the efficacy of the design algorithm. Focusing on the interaction between the SARS-CoV-2 Spike Receptor-Binding Domain (RBD) and the human angiotensin-converting enzyme 2 (ACE2) receptor, we extracted a 23-residues long peptide that structurally mimics the major interacting portion of the ACE2 receptor and designed in silico five mutants of such a peptide with a modulated affinity. Remarkably, experimental KD measurements, conducted using biolayer interferometry, matched the in silico predictions. Moreover, we investigated the molecular determinants that govern the variation in binding affinity through molecular dynamics simulation, by identifying the mechanisms driving the different values of binding affinity at a single residue level. Finally, the peptide sequence with the highest affinity, in comparison with the wild type peptide, was expressed as a fusion protein with human H ferritin (HFt) 24-mer. Solution measurements performed on the latter constructs confirmed that peptides still exhibited the expected trend, thereby enhancing their efficacy in RBD binding. Altogether, these results indicate the high potentiality of this general method in developing potent high-affinity vectors for hindering/enhancing protein-protein associations.

Smart design of universally decorated nanoparticles for drug delivery applications driven by active transport.

Targeting the cell nucleus remains a challenge for drug delivery. Here, we present a universal platform for the smart design of nanoparticle (NP) decoration that is based on: (i) a spacer polymer, commonly biotin-polyethylene-glycol-thiol, whose grafting density and molecular weight can be tuned for optimized performance, and (ii) protein binding peptides, such as cell penetrating peptides (CPPs), cancer-targeting peptides, or nuclear localization signal (NLS) peptides, that are linked to the PEG free-end by universal chemistry. We manifested our platform with two different bromo-acetamide (Br-Ac) modified NLSs. We used cell extract-based and live cell assays to demonstrate the recruitment of dynein motor proteins, which drive the NP active transport toward the nucleus, and the enhancement of cellular and nuclear entry, manifesting the properties of NLS as a CPP. Our control of the NP decoration scheme, and the modularity of our platform, carry great advantages for nano-carrier design for drug delivery applications.

Sequential intervention of anti-inflammatory and osteogenesis with silk fibroin coated polyethylene terephthalate artificial ligaments for anterior cruciate ligament reconstruction.

Graft-host integration after the anterior cruciate ligament (ACL) reconstruction sequentially follows the prognosis from the inflammation period to the regeneration period. However, due to insufficient bioactivity, polyethylene terephthalate (PET) artificial ligaments often require a long period for graft-host integration. To improve graft-host integration, sequential therapy targeting multifactor is widely advocated. In this study, a multilayer regenerated silk fibroin (RSF) coating loaded with heparin and bone morphogenetic protein binding peptide (BBP) for differentiated release was introduced on the surface of the PET artificial ligament by a stepwise deposition method. The drug release profiles of heparin and BBP on the coated PET artificial ligament indicated the features of differential drug release, i.e., with heparin in the outermost layer releasing a significant amount (more than 60%) during the first 5 days while BBP in the inner layer only releasing a small amount (ca. 30%) within 1 week without burst release. Based on the isometric ACL reconstruction model of rabbits, such drug-loaded RSF coating was verified to be able to modulate the early inflammatory response and promote the maturation of the graft in the articular cavity, meanwhile, it provided a continuous and stable signal of osteogenic induction to improve graft-bone integration. Thus, sequential intervention with heparin and BBP proved to be a reliable combination, and multifunctional RSF-coated PET artificial ligaments hold great potential for improving the clinical efficacy of ACL reconstruction.

Dimeric translationally controlled tumor protein-binding peptide 2 attenuates imiquimod-induced psoriatic inflammation through induction of regulatory T cells.

Psoriasis is a chronic skin inflammation caused by a dysfunctional immune system, which causes systemic inflammation in various organs and tissues. Due to the risk of systemic inflammation and recurrence of psoriasis, it is important to identify the critical targets in the pathogenesis of psoriasis and develop targeted therapeutics. Dimerized translationally controlled tumor protein (dTCTP) promotes immune cell activation as a pro-inflammatory cytokine and plays a role in developing allergic diseases such as asthma and rhinitis. Here, we sought to explore whether dTCTP and its inhibition contributed to the development and control of imiquimod (IMQ)-induced psoriasis. Topical application of IMQ inflamed the skin of the back and ear, increased inflammatory cytokines, and decreased regulatory T cell markers. Interestingly, TCTP was significantly increased in inflamed skin and immune cells such as T cells, B cells, and macrophages after IMQ treatment and was secreted into the serum to undergo dimerization. Extracellular dTCTP treatment selectively suppressed regulatory T (Treg) cells, not other effector T helper (Th) cells, and increased M1 macrophages. Moreover, dTCTP-binding peptide 2 (dTBP2), a dTCTP inhibitor peptide, effectively attenuated the systemic inflammatory responses, including Th17 cell response, and alleviated psoriatic skin inflammation. dTBP2 blocked dTCTP-mediated Treg suppression and stimulated the expression of Treg cell markers in the spleen and inflammatory skin lesions. These results suggest that dTCTP dysregulated immune balance through Treg suppression in psoriatic inflammation and that functional inhibition of dTCTP by dTBP2 maintained immune homeostasis and attenuated inflammatory skin diseases by expanding Treg cells.

Targeting the Surface of the Protein 14-3-3 by Ultrasmall (1.5 nm) Gold Nanoparticles Carrying the Specific Peptide CRaf.

The surface of ultrasmall gold nanoparticles with an average diameter of 1.55 nm was conjugated with a 14‐3‐3 protein‐binding peptide derived from CRaf. Each particle carries 18 CRaf peptides, leading to an overall stoichiometry of Au(115)Craf(18). The binding to the protein 14‐3‐3 was probed by isothermal titration calorimetry (ITC) and fluorescence polarization spectroscopy (FP). The dissociation constant (K D) was measured as 5.0 μM by ITC and 0.9 μM by FP, which was close to the affinity of dissolved CRaf to 14‐3‐3σ. In contrast to dissolved CRaf, which alone did not enter HeLa cells, CRAF‐conjugated gold nanoparticles were well taken up by HeLa cells, opening the opportunity to target the protein inside a cell.

HDM2 protein-binding peptides screened from stapled α-helical peptide phage display libraries with different types of staple linkers

Chemically modified peptide ligands were identified from α-helix peptide phage libraries with different types of staple linkers. The hDM2-protein was used as a representative target of protein–protein interactions to screen ligands for p53 binding sites in hDM2. Two types of staple linkers were used for the chemical modification of the peptide phage display libraries before affinity selection. The identified stapled peptides could bind to hDM2 competitively with the p53 peptide. The stapled peptide phage libraries developed in this study will improve the discovery of protein–protein interaction inhibitors through the synergistic effect of peptide units and staple linkers.

Translating antibody-binding peptides into peptoid ligands with improved affinity and stability

A great number of protein-binding peptides are known and utilized as drugs, diagnostic reagents, and affinity ligands. Recently, however, peptide mimetics have been proposed as valuable alternative to peptides by virtue of their excellent biorecognition activity and higher biochemical stability. This poses the need to develop a strategy for translating known protein-binding peptides into peptoid analogues with comparable or better affinity. This work proposes a route for translation utilizing the IgG-binding peptide HWRGWV as reference sequence. An ensemble of peptoid analogues of HWRGWV were produced by adjusting the number and sequence arrangement of residues containing functional groups that resemble both natural and non-natural amino acids. The variants were initially screened via IgG binding tests in non-competitive mode to select candidate ligands. A set of selected peptoids were studied in silico by docking onto putative binding sites identified on the crystal structures of human IgG1, IgG2, IgG3, and IgG4 subclasses, returning values of predicted binding energy that aligned well with the binding data. Selected peptoids PL-16 and PL-22 were further characterized by binding isotherm analysis to determine maximum capacity (Qmax ˜ 48–57 mg of IgG per mL of adsorbent) and binding strength on solid phase (KD ˜ 5.4–7.8 10−7 M). Adsorbents PL-16-Workbeads and PL-22-Workbeads were used for purifying human IgG from a cell culture supernatant added with bovine serum, affording high values of IgG recovery (up to 85%) and purity (up to 98%) under optimized binding and elution conditions. Both peptoid ligands also proved to be stable against proteolytic enzymes and strong alkaline agents. Collectively, these studies form a method guiding the design of peptoid variants of cognate peptide ligands, and help addressing the challenges that, despite the structural similarity, the peptide-to-peptoid translation presents.

Peptide-Directed DNA-Templated Protein Labelling for The Assembly of a Pseudo-IgM.

The development of methods for conjugation of DNA to proteins is of high relevance for the integration of protein function and DNA structures. Here, we demonstrate that protein-binding peptides can direct a DNA-templated reaction, selectively furnishing DNA-protein conjugates with one DNA label. Quantitative conversion of oligonucleotides is achieved at low stoichiometries and the reaction can be performed in complex biological matrixes, such as cell lysates. Further, we have used a star-like pentameric DNA nanostructure to assemble five DNA-Rituximab conjugates, made by our reported method, into a pseudo-IgM antibody structure that was subsequently characterized by negative-stain transmission electron microscopy (nsTEM) analysis.

Nonproteinogenic deep mutational scanning of linear and cyclic peptides

High-resolution structure-activity analysis of polypeptides requires amino acid structures that are not present in the universal genetic code. Examination of peptide and protein interactions with this resolution has been limited by the need to individually synthesize and test peptides containing nonproteinogenic amino acids. We describe a method to scan entire peptide sequences with multiple nonproteinogenic amino acids and, in parallel, determine the thermodynamics of binding to a partner protein. By coupling genetic code reprogramming to deep mutational scanning, any number of amino acids can be exhaustively substituted into peptides, and single experiments can return all free energy changes of binding. We validate this approach by scanning two model protein-binding peptides with 21 diverse nonproteinogenic amino acids. Dense structure-activity maps were produced at the resolution of single aliphatic atom insertions and deletions. This permits rapid interrogation of interaction interfaces, as well as optimization of affinity, fine-tuning of physical properties, and systematic assessment of nonproteinogenic amino acids in binding and folding.

Encoding Stem-Cell-Secreted Extracellular Matrix Protein Capture in Two and Three Dimensions Using Protein Binding Peptides

Capturing cell-secreted extracellular matrix (ECM) proteins through cooperative binding with high specificity and affinity is an important function of native tissue matrices during both tissue homeostasis and repair. However, while synthetic hydrogels, such as those based on poly(ethylene glycol) (PEG), are often proposed as ideal materials to deliver human mesenchymal stem cells (hMSCs) to sites of injury to enable tissue repair, they do not have this capability-a capability that would enable cells to actively remodel their local extracellular microenvironment and potentially provide the required feedback control for more effective tissue genesis. In this work, we detail a methodology that engenders poly(ethylene glycol) (PEG)-based two-dimensional substrates and three-dimensional porous hydrogels with the ability to capture desired extracellular matrix (ECM) proteins with high specificity. This "encoded" ECM protein capture is achieved by decorating the PEG-based materials with protein binding peptides (PBPs) synthesized to be specific in their binding of fibronectin, laminin, and collagen I, which are not only the most omnipresent ECM proteins in human tissues but, as we confirmed, are also secreted to differing extents by hMSCs under in vitro maintenance conditions. By encapsulating hMSCs into these PBP-functionalized hydrogels, and culturing them in protein-free maintenance media, we demonstrate that these PBPs not only actively recruit targeted ECM proteins as they are secreted from hMSCs but also retain them to much higher levels compared to nonfunctionalized gels. This novel approach thus enables the fabrication of encoded surfaces and hydrogels that capture cell-secreted proteins, with high specificity and affinity, in a programmable manner, ready for applications in many bioengineering applications, including bioactive surface coatings, bioassays, stem cell culture, tissue engineering, and regenerative medicine.

Turn-on fluorescence detection of protein by molecularly imprinted hydrogels based on supramolecular assembly of peptide multi-functional blocks.

Synthetic receptors for biomacromolecules lack the supramolecular self-assembly behavior typical of biological systems. Here we propose a new method for the preparation of protein imprinted polymers based on the specific interaction of a peptide multi-functional block with a protein target. This peptide block contains a protein-binding peptide domain, a polymerizable moiety at the C-terminus and an environment-sensitive fluorescent molecule at the N-terminus. The method relies on a preliminary step consisting of peptide/protein supramolecular assembly, followed by copolymerization with the most common acrylate monomers (acrylamide, acrylic acid and bis-acrylamide) to produce a protein imprinted hydrogel polymer. Such a peptide block can function as an active assistant recognition element to improve affinity, and guarantees its effective polymerization at the protein/cavity interface, allowing for proper placement of a dye. As a proof of concept, we chose Bovine Serum Albumin (BSA) as the protein target and built the peptide block around a BSA binding dodecapeptide, with an allyl group as the polymerizable moiety and a dansyl molecule as the responsive dye. Compared to conventional approaches these hydrogels showed higher affinity (more than 45%) and imprinted sensitivity (about twenty fold) to the target, with a great BSA selectivity with respect to ovalbumin (α = 1.25) and lysozyme (α = 6.02). Upon protein binding, computational and experimental observations showed a blue shift of the emission peak (down to 440 nm) and an increase of fluorescence emission (twofold) and average lifetime (Δτ = 4.3 ns). Such an approach generates recognition cavities with controlled chemical information and represents an a priori method for self-responsive materials. Provided a specific peptide and minimal optimization conditions are used, such a method could be easily implemented for any protein target.

Dimerized Translationally Controlled Tumor Protein-Binding Peptide Ameliorates Atopic Dermatitis in NC/Nga Mice.

Our previous study showed that dimerized translationally controlled tumor protein (dTCTP) plays a role in the pathogenesis of allergic diseases, such as asthma and allergic rhinitis. A 7-mer peptide, called dTCTP-binding peptide 2 (dTBP2), binds to dTCTP and inhibits its cytokine-like effects. We therefore examined the protective effects of dTBP2 in house dust mite-induced atopic dermatitis (AD)-like skin lesions in Nishiki-nezumi Cinnamon/Nagoya (NC/Nga) mice. We found that topical administration of dTBP2 significantly reduced the AD-like skin lesions formation and mast cell infiltration in NC/Nga mice, similarly to the response seen in the Protopic (tacrolimus)-treated group. Treatment with dTBP2 also decreased the serum levels of IgE and reduced IL-17A content in skin lesions and inhibited the expression of mRNAs of interleukin IL-4, IL-5, IL-6, IL-13, macrophage-derived chemokine (MDC), thymus and activation-regulated chemokine (TARC) and thymic stromal lymphopoietin (TSLP). These findings indicate that dTBP2 not only inhibits the release of Th2 cytokine but also suppresses the production of proinflammatory cytokines in AD-like skin lesions in NC/Nga mice, by inhibiting TCTP dimer, in allergic responses. Therefore, dTCTP is a therapeutic target for AD and dTBP2 appears to have a potential role in the treatment of AD.

Cell-Permeable Peptides Containing Cycloalanine Residues.

We present here an efficient alternative to N-methylation for the purpose of morphing protein-binding peptides into more serum-stable and cell-permeable compounds. This involves the incorporation of a cycloalanine (CyAla) into a peptide in a way that avoids difficult coupling steps. We demonstrate the utility of this chemistry in creating a cell-permeable derivative of a high-affinity HIV Rev protein-binding peptide.

Proteolysis targeting peptide (PROTAP) strategy for protein ubiquitination and degradation

Ubiquitination proteasome pathway (UPP) is the most important and selective way to degrade proteins in vivo. Here, a novel proteolysis targeting peptide (PROTAP) strategy, composed of a target protein binding peptide, a linker and a ubiquitin E3 ligase recognition peptide, was designed to recruit both target protein and E3 ligase and then induce polyubiquitination and degradation of the target protein through UPP. In our study, the PROTAP strategy was proved to be a general method with high specificity using Bcl-xL protein as model target in vitro and in cells, which indicates that the strategy has great potential for in vivo application.

Peptides and Peptide Analogs to Inhibit Protein-Protein Interactions.

Protein-protein interactions are governed by relatively few amino acid residues at the binding interface. Peptides derived from these protein regions may serve as mimics of one of the interaction partners in structural studies or as inhibitors to disrupt the respective interaction and investigate its biological consequences. Inhibitory peptides may also be lead structures for drug development if the respective protein-protein interaction is essential for a pathogen or disease mechanism. Binding peptides may be systematically derived from one of the binding partners or found in the screen of combinatorial peptide libraries. Molecular modelling based on structural data helps to refine existing peptides or even design novel binding peptides. This chapter gives an outline of the binding peptide discovery process and subsequent chemical modifications to further enhance affinity and specificity and to increase stability against degradation in vivo. Examples from the past three decades illustrate the great diversity of applications for protein binding peptides and peptide analogs.

A purification method for a molecular complex in which a scaffold molecule is fully loaded with heterogeneous molecules.

An affinity resin-based pull-down method is convenient for the purification of biochemical materials. However, its use is difficult for the isolation of a molecular complex fully loaded with multiple components from a reaction mixture containing the starting materials and intermediate products. To overcome this problem, we have developed a new purification procedure that depends on sequential elimination of the residues. In practice, two affinity resins were used for purifying a triangular-shaped RNP (RNA-protein complex) consisting of three ribosomal proteins (L7Ae) bound to an RNA scaffold. First, a resin with immobilized L7Ae protein captured the incomplete RNP complexes and the free RNA scaffold. Next, another resin with an immobilized chemically modified RNA of a derivative of Box C/D motif, the binding partner of L7Ae, was used to capture free protein. The complete triangular RNP was successfully purified from the mixture by these two steps. Obviously, the purified triangular RNP displaying three protein-binding peptides exhibited an improved performance when compared with the unrefined product. Conceptually, this purification procedure should be applicable for the purification of a variety of complexes consisting of multiple components other than RNP.

Virtual screening using combinatorial cyclic peptide libraries reveals protein interfaces readily targetable by cyclic peptides.

Protein-protein and protein-peptide interactions are responsible for the vast majority of biological functions in vivo, but targeting these interactions with small molecules has historically been difficult. What is required are efficient combined computational and experimental screening methods to choose among a number of potential protein interfaces worthy of targeting lead macrocyclic compounds for further investigation. To achieve this, we have generated combinatorial 3D virtual libraries of short disulfide-bonded peptides and compared them to pharmacophore models of important protein-protein and protein-peptide structures, including short linear motifs (SLiMs), protein-binding peptides, and turn structures at protein-protein interfaces, built from 3D models available in the Protein Data Bank. We prepared a total of 372 reference pharmacophores, which were matched against 108,659 multiconformer cyclic peptides. After normalization to exclude nonspecific cyclic peptides, the top hits notably are enriched for mimetics of turn structures, including a turn at the interaction surface of human α thrombin, and also feature several protein-binding peptides. The top cyclic peptide hits also cover the critical "hot spot" interaction sites predicted from the interaction crystal structure. We have validated our method by testing cyclic peptides predicted to inhibit thrombin, a key protein in the blood coagulation pathway of important therapeutic interest, identifying a cyclic peptide inhibitor with lead-like activity. We conclude that protein interfaces most readily targetable by cyclic peptides and related macrocyclic drugs may be identified computationally among a set of candidate interfaces, accelerating the choice of interfaces against which lead compounds may be screened.