Peptidic modulators of protein-protein interactions: progress and challenges in computational design

GraphPBSP: Protein binding site prediction based on Graph Attention Network and pre-trained model ProstT5

Protein-protein/peptide interactions play crucial roles in various biological processes. Exploring their interactions attracts wide attention. However, accurately predicting their binding sites remains a challenging task. Here, we develop an effective model GraphPBSP based on Graph Attention Network with Convolutional Neural Network and Multilayer Perceptron for protein-protein/peptide binding site prediction, which utilizes various feature types derived from protein sequence and structure including interface residue pairwise propensity developed by us and sequence embeddings obtained from a new pre-trained model ProstT5, alongside physicochemical properties and structural features. To our best knowledge, ProstT5 sequence embeddings and residue pairwise propensity are first introduced for protein-protein/peptide binding site prediction. Additionally, we propose a spatial neighbor-based feature statistic method for effectively considering key spatially neighboring information that significantly improves the model's prediction ability. For model training, a multi-scale objective function is constructed, which enhances the learning capability across samples of the same or different classes. On multiple protein-protein/peptide binding site test sets, GraphPBSP outperforms the currently available state-of-the-art methods with an excellent performance. Additionally, its performances on protein-DNA/RNA binding site test sets also demonstrate its good generalization ability. In conclusion, GraphPBSP is a promising method, which can offer valuable information for protein engineering and drug design.

Protein-peptide binding residue prediction based on protein language models and cross-attention mechanism

Accurate identifications of protein-peptide binding residues are essential for protein-peptide interactions and advancing drug discovery. To address this problem, extensive research efforts have been made to design more discriminative feature representations. However, extracting these explicit features usually depend on third-party tools, resulting in low computational efficacy and suffering from low predictive performance. In this study, we design an end-to-end deep learning-based method, E2EPep, for protein-peptide binding residue prediction using protein sequence only. E2EPep first employs and fine-tunes two state-of-the-art pre-trained protein language models that can extract two different high-latent feature representations from protein sequences relevant for protein structures and functions. A novel feature fusion module is then designed in E2EPep to fuse and optimize the above two feature representations of binding residues. In addition, we have also design E2EPep+, which integrates E2EPep and PepBCL models, to improve the prediction performance. Experimental results on two independent testing data sets demonstrate that E2EPep and E2EPep + could achieve the average AUC values of 0.846 and 0.842 while achieving an average Matthew's correlation coefficient value that is significantly higher than that of existing most of sequence-based methods and comparable to that of the state-of-the-art structure-based predictors. Detailed data analysis shows that the primary strength of E2EPep lies in the effectiveness of feature representation using cross-attention mechanism to fuse the embeddings generated by two fine-tuned protein language models. The standalone package of E2EPep and E2EPep + can be obtained at https://github.com/ckx259/E2EPep.git for academic use only.

PepCA: Unveiling protein-peptide interaction sites with a multi-input neural network model

The protein-peptide interaction plays a pivotal role in fields such as drug development, yet remains underexplored experimentally and challenging to model computationally. Herein, we introduce PepCA, a sequence-based approach for predicting peptide-binding sites on proteins. A primary obstacle in predicting peptide-protein interactions is the difficulty in acquiring precise protein structures, coupled with the uncertainty of polypeptide configurations. To address this, we first encode protein sequences using the Evolutionary Scale Modeling 2 (ESM-2) pre-trained model to extract latent structural information. Additionally, we have developed a multi-input coattention mechanism to concurrently update the encoding of both peptide and protein residues. PepCA integrates this module within an encoder-decoder structure. This model's high precision in identifying binding sites significantly advances the field of computational biology, offering vital insights for peptide drug development and protein science.

MucLiPred: Multi-Level Contrastive Learning for Predicting Nucleic Acid Binding Residues of Proteins

Protein-molecule interactions play a crucial role in various biological functions, with their accurate prediction being pivotal for drug discovery and design processes. Traditional methods for predicting protein-molecule interactions are limited. Some can only predict interactions with a specific molecule, restricting their applicability, while others target multiple molecule types but fail to efficiently process diverse interaction information, leading to complexity and inefficiency. This study presents a novel deep learning model, MucLiPred, equipped with a dual contrastive learning mechanism aimed at improving the prediction of multiple molecule-protein interactions and the identification of potential molecule-binding residues. The residue-level paradigm focuses on differentiating binding from non-binding residues, illuminating detailed local interactions. The type-level paradigm, meanwhile, analyzes overarching contexts of molecule types, like DNA or RNA, ensuring that representations of identical molecule types gravitate closer in the representational space, bolstering the model's proficiency in discerning interaction motifs. This dual approach enables comprehensive multi-molecule predictions, elucidating the relationships among different molecule types and strengthening precise protein-molecule interaction predictions. Empirical evidence demonstrates MucLiPred's superiority over existing models in robustness and prediction accuracy. The integration of dual contrastive learning techniques amplifies its capability to detect potential molecule-binding residues with precision. Further optimization, separating representational and classification tasks, has markedly improved its performance. MucLiPred thus represents a significant advancement in protein-molecule interaction prediction, setting a new precedent for future research in this field.

DeepProSite: structure-aware protein binding site prediction using ESMFold and pretrained language model

MOTIVATION

Identifying the functional sites of a protein, such as the binding sites of proteins, peptides, or other biological components, is crucial for understanding related biological processes and drug design. However, existing sequence-based methods have limited predictive accuracy, as they only consider sequence-adjacent contextual features and lack structural information.

RESULTS

In this study, DeepProSite is presented as a new framework for identifying protein binding site that utilizes protein structure and sequence information. DeepProSite first generates protein structures from ESMFold and sequence representations from pretrained language models. It then uses Graph Transformer and formulates binding site predictions as graph node classifications. In predicting protein-protein/peptide binding sites, DeepProSite outperforms state-of-the-art sequence- and structure-based methods on most metrics. Moreover, DeepProSite maintains its performance when predicting unbound structures, in contrast to competing structure-based prediction methods. DeepProSite is also extended to the prediction of binding sites for nucleic acids and other ligands, verifying its generalization capability. Finally, an online server for predicting multiple types of residue is established as the implementation of the proposed DeepProSite.

AVAILABILITY AND IMPLEMENTATION

The datasets and source codes can be accessed at https://github.com/WeiLab-Biology/DeepProSite. The proposed DeepProSite can be accessed at https://inner.wei-group.net/DeepProSite/.

PepCNN deep learning tool for predicting peptide binding residues in proteins using sequence, structural, and language model features

Protein-peptide interactions play a crucial role in various cellular processes and are implicated in abnormal cellular behaviors leading to diseases such as cancer. Therefore, understanding these interactions is vital for both functional genomics and drug discovery efforts. Despite a significant increase in the availability of protein-peptide complexes, experimental methods for studying these interactions remain laborious, time-consuming, and expensive. Computational methods offer a complementary approach but often fall short in terms of prediction accuracy. To address these challenges, we introduce PepCNN, a deep learning-based prediction model that incorporates structural and sequence-based information from primary protein sequences. By utilizing a combination of half-sphere exposure, position specific scoring matrices from multiple-sequence alignment tool, and embedding from a pre-trained protein language model, PepCNN outperforms state-of-the-art methods in terms of specificity, precision, and AUC. The PepCNN software and datasets are publicly available at https://github.com/abelavit/PepCNN.git .

Integrated unsupervised-supervised modeling and prediction of protein-peptide affinities at structural level

Cell signal networks are orchestrated directly or indirectly by various peptide-mediated protein-protein interactions, which are normally weak and transient and thus ideal for biological regulation and medicinal intervention. Here, we develop a general-purpose method for modeling and predicting the binding affinities of protein-peptide interactions (PpIs) at the structural level. The method is a hybrid strategy that employs an unsupervised approach to derive a layered PpI atom-residue interaction (ulPpI[a-r]) potential between different protein atom types and peptide residue types from thousands of solved PpI complex structures and then statistically correlates the potential descriptors with experimental affinities (KD values) over hundreds of known PpI samples in a supervised manner to create an integrated unsupervised-supervised PpI affinity (usPpIA) predictor. Although both the ulPpI[a-r] potential and usPpIA predictor can be used to calculate PpI affinities from their complex structures, the latter seems to perform much better than the former, suggesting that the unsupervised potential can be improved substantially with a further correction by supervised statistical learning. We examine the robustness and fault-tolerance of usPpIA predictor when applied to treat the coarse-grained PpI complex structures modeled computationally by sophisticated peptide docking and dynamics simulation. It is revealed that, despite developed solely based on solved structures, the integrated unsupervised-supervised method is also applicable for locally docked structures to reach a quantitative prediction but can only give a qualitative prediction on globally docked structures. The dynamics refinement seems not to change (or improve) the predictive results essentially, although it is computationally expensive and time-consuming relative to peptide docking. We also perform extrapolation of usPpIA predictor to the indirect affinity quantities of HLA-A*0201 binding epitope peptides and NHERF PDZ binding scaffold peptides, consequently resulting in a good and moderate correlation of the predicted KD with experimental IC50 and BLU on the two peptide sets, with Pearson's correlation coefficients Rp = 0.635 and 0.406, respectively.

Inhibition of Viral Membrane Fusion by Peptides and Approaches to Peptide Design

Fusion of lipid-enveloped viruses with the cellular plasma membrane or the endosome membrane is mediated by viral envelope proteins that undergo large conformational changes following binding to receptors. The HIV-1 fusion protein gp41 undergoes a transition into a "six-helix bundle" after binding of the surface protein gp120 to the CD4 receptor and a co-receptor. Synthetic peptides that mimic part of this structure interfere with the formation of the helix structure and inhibit membrane fusion. This approach also works with the S spike protein of SARS-CoV-2. Here we review the peptide inhibitors of membrane fusion involved in infection by influenza virus, HIV-1, MERS and SARS coronaviruses, hepatitis viruses, paramyxoviruses, flaviviruses, herpesviruses and filoviruses. We also describe recent computational methods used for the identification of peptide sequences that can interact strongly with protein interfaces, with special emphasis on SARS-CoV-2, using the PePI-Covid19 database.

Improving Protein-Peptide Docking Results via Pose-Clustering and Rescoring with a Combined Knowledge-Based and MM-GBSA Scoring Function

Protein-peptide docking, which predicts the complex structure between a protein and a peptide, is a valuable computational tool in peptide therapeutics development and the mechanistic investigation of peptides involved in cellular processes. Although current peptide docking approaches are often able to sample near-native peptide binding modes, correctly identifying those near-native modes from decoys is still challenging because of the extremely high complexity of the peptide binding energy landscape. In this study, we have developed an efficient postdocking rescoring protocol using a combined scoring function of knowledge-based ITScorePP potentials and physics-based MM-GBSA energies. Tested on five benchmark/docking test sets, our postdocking strategy showed an overall significantly better performance in binding mode prediction and score-rmsd correlation than original docking approaches. Specifically, our postdocking protocol outperformed original docking approaches with success rates of 15.8 versus 10.5% for pepATTRACT on the Global_57 benchmark, 5.3 versus 5.3% for CABS-dock on the Global_57 benchmark, 17.0 versus 11.3% for FlexPepDock on the LEADS-PEP data set, 40.3 versus 33.9% for HPEPDOCK on the Local_62 benchmark, and 64.2 versus 52.8% for HPEPDOCK on the LEADS-PEP data set when the top prediction was considered. These results demonstrated the efficacy and robustness of our postdocking protocol.

Proteasome Inhibitor Drugs

Proteasomes are large, multicatalytic protein complexes that cleave cellular proteins into peptides. There are many distinct forms of proteasomes that differ in catalytically active subunits, regulatory subunits, and associated proteins. Proteasome inhibitors are an important class of drugs for the treatment of multiple myeloma and mantle cell lymphoma, and they are being investigated for other diseases. Bortezomib (Velcade) was the first proteasome inhibitor to be approved by the US Food and Drug Administration. Carfilzomib (Kyprolis) and ixazomib (Ninlaro) have recently been approved, and more drugs are in development. While the primary mechanism of action is inhibition of the proteasome, the downstream events that lead to selective cell death are not entirely clear. Proteasome inhibitors have been found to affect protein turnover but at concentrations that are much higher than those achieved clinically, raising the possibility that some of the effects of proteasome inhibitors are mediated by other mechanisms.

Molecular dynamics insights on the role β-augmentation of the peptide N-terminus with binding site β-hairpin of proprotein convertase subtilisin/kexin 9

PCSK9, a member of the proprotein convertase family, is a key negative regulator of hepatic low-density lipoprotein receptor (LDLR) concentrations in the blood plasma and is associated with the risk of coronary artery disease (CAD). Peptide inhibitors designed to block PCSK9-LDLR interactions could reduce the risk of CAD. We present a study of the interaction of a PCSK9 bound peptide and its design through modification by phosphorylation using molecular dynamics simulations. Extensive explicit solvent simulations of PCSK9 and its mutant (Asp374 → Tyr374) with designed peptides provide insights into the mechanism of peptide binding at the protein interface. We establish that β-augmentation is the key mechanism of peptide association with PCSK9. Position-specific phosphorylation of threonine residues is observed to have noticeable effect in modulating protein-peptide association. This study provides a handle to explore and improve the design of peptides bound to PCSK9 by incorporating knowledge-derived functional motifs into designing potent binders.

Machine Learning in Quantitative Protein–peptide Affinity Prediction: Implications for Therapeutic Peptide Design

Background:

Protein–peptide recognition plays an essential role in the orchestration and regulation of cell signaling networks, which is estimated to be responsible for up to 40% of biological interaction events in the human interactome and has recently been recognized as a new and attractive druggable target for drug development and disease intervention.

Methods:

We present a systematic review on the application of machine learning techniques in the quantitative modeling and prediction of protein–peptide binding affinity, particularly focusing on its implications for therapeutic peptide design. We also briefly introduce the physical quantities used to characterize protein–peptide affinity and attempt to extend the content of generalized machine learning methods.

Results:

Existing issues and future perspective on the statistical modeling and regression prediction of protein– peptide binding affinity are discussed.

Conclusion:

There is still a long way to go before establishment of general, reliable and efficient machine leaningbased protein–peptide affinity predictors.

Influence of Amino Acid Mutations and Small Molecules on Targeted Inhibition of Proteins Involved in Cancer

Background:

Protein-protein interactions (PPIs) are of crucial importance in regulating the biological processes of cells both in normal and diseased conditions. Significant progress has been made in targeting PPIs using small molecules and achieved promising results. However, PPI drug discovery should be further accelerated with better understanding of chemical space along with various functional aspects.

Objective:

In this review, we focus on the advancements in computational research for targeted inhibition of protein-protein interactions involved in cancer.

Methods:

Here, we mainly focused on two aspects: (i) understanding the key roles of amino acid mutations in epidermal growth factor receptor (EGFR) as well as mutation-specific inhibitors and (ii) design of small molecule inhibitors for Bcl-2 to disrupt PPIs.

Results:

The paradigm of PPI inhibition to date reflect the certainty that inclination towards novel and versatile strategies enormously dictate the success of PPI inhibition. As the chemical space highly differs from the normal drug like compounds the lead optimization process has to be given the utmost priority to ensure the clinical success. Here, we provided a broader perspective on effect of mutations in oncogene EGFR connected to Bcl-2 PPIs and focused on the potential challenges.

Conclusion:

Understanding and bridging mutations and altered PPIs will provide insights into the alarming signals leading to massive malfunctioning of a biological system in various diseases. Finding rational elucidations from a pharmaceutical stand point will presumably broaden the horizons in future.

Identification of a peptide derived from a Bothrops moojeni metalloprotease with in vitro inhibitory action on the Plasmodium falciparum purine nucleoside phosphorylase enzyme (PfPNP)

There is a growing need for research on new antimalarial agents against Plasmodium falciparum infection, especially in regards to planning molecular architecture for specific molecular targets of the parasite. Thus, a metalloprotease from Bothrops moojeni, known as BmooMPα-I, was explored in this study, through in silico assays, aiming at the development of a peptide generated from this molecule with potential inhibitory action on PfPNP, an enzyme necessary for the survival of the parasite. In order to isolate BmooMPα-I, cation exchange and reverse phase chromatographies were performed, followed by in vitro assays of antiparasitic activity against the W2 strain of P. falciparum. The interactions between BmooMPα-I and PfPNP were evaluated via docking, and the resulting peptide, described as Pep1 BM, was selected according to the BmooMPα-I region demonstrating the best interaction score with the target of interest. The values for the specific activities of the PfPNP reaction were measured using the inorganic phosphate substrate and MESG. The fraction corresponding to BmooMPα-I was identified as fraction 4 in the cation exchange chromatography step, due to proteolytic activity on casein and the presence of a major band at ≅ 23 kDa. BmooMPα-I was able to inhibit in vitro growth of W2 P. falciparum, with an IC₅₀ value of 16.14 μg/mL. Virtual screening with Pep1 BM demonstrated two PfPNP target binding regions, with ΔG values at the interaction interface of -10.75 kcal/mol and -11.74 kcal/mol. A significant reduction in the enzymatic activity of PfPNP was observed in the presence of Pep 1 BM when compared to the assay in the absence of this possible inhibitor. BmooMPα-I showed activity in vitro against W2 P. falciparum. By means of in silico techniques, the Pep 1 BM was identified as having potential binding affinity to the catalytic site of PfPNP and of inhibiting its catalytic activity in vitro.

In Silico Peptide Ligation: Iterative Residue Docking and Linking as a New Approach to Predict Protein-Peptide Interactions

Peptide⁻protein interactions are corner-stones of living functions involved in essential mechanisms, such as cell signaling. Given the difficulty of obtaining direct experimental structural biology data, prediction of those interactions is of crucial interest for the rational development of new drugs, notably to fight diseases, such as cancer or Alzheimer's disease. Because of the high flexibility of natural unconstrained linear peptides, prediction of their binding mode in a protein cavity remains challenging. Several theoretical approaches have been developed in the last decade to address this issue. Nevertheless, improvements are needed, such as the conformation prediction of peptide side-chains, which are dependent on peptide length and flexibility. Here, we present a novel in silico method, Iterative Residue Docking and Linking (IRDL), to efficiently predict peptide⁻protein interactions. In order to reduce the conformational space, this innovative method splits peptides into several short segments. Then, it uses the performance of intramolecular covalent docking to rebuild, sequentially, the complete peptide in the active site of its protein target. Once the peptide is constructed, a rescoring step is applied in order to correctly rank all IRDL solutions. Applied on a set of 11 crystallized peptide⁻protein complexes, the IRDL method shows promising results, since it is able to retrieve experimental binding conformations with a Root Mean Square Deviation (RMSD) below 2 Å in the top five ranked solutions. For some complexes, IRDL method outperforms two other docking protocols evaluated in this study. Hence, IRDL is a new tool that could be used in drug design projects to predict peptide⁻protein interactions.

In Silico Approaches for TRP Channel Modulation

The implication of several TRP ion channels (e.g., TRPV1) in diverse physiological and pathological processes has signaled them as pivotal drug targets. Consequently, the identification of selective and potent ligands for these channels is of great interest in pharmacology and biomedicine. However, a major challenge in the design of modulators is ensuring the specificity for their intended targets. In recent years, the emergence of high-resolution structures of ion channels facilitates the computer-assisted drug design at molecular levels. Here we describe some computational methods and general protocols to discover channel modulators, including homology modelling, docking and virtual screening, and structure-based peptide design.

Computational studies on ground and excited state charge transfer properties of peptidomimetics

Chemical modifications at various peptide positions result in peptidomimetics with unique physical and chemical properties that can be used for a range of applications. Among many peptidomimetics, ureidopeptides are interesting due to their ability to act as donor-bridge-acceptor systems through which charge transfer occurs in one direction and can be triggered by an electrochemical pulse without perturbing the nuclear position. In this regard, some UP mimetics with different chromophoric units are studied in this work to understand their role using DFT based methods. Computational results and natural charge analysis provide evidence for the extensive contribution of the substituents to the excitation and hole migration dynamics. Further, the results show that the UP backbone preserves its uni-directional charge transfer phenomenon from the ureido to carboxylate terminal irrespective of the terminal groups and position. However, the substituent affects the excitation energies and the time scales of the hole migration. Among the substituents studied here, fluorine migrates to the hole within a shorter time scale while phenyl groups take longer.

Application of the ATTRACT Coarse-Grained Docking and Atomistic Refinement for Predicting Peptide-Protein Interactions

Peptide-protein interactions are abundant in the cell and form an important part of the interactome. Large-scale modeling of peptide-protein complexes requires a fully blind approach; i.e., simultaneously predicting the peptide-binding site and the peptide conformation to high accuracy. Here, we present one of the first fully blind peptide-protein docking protocols, pepATTRACT. It combines a coarse-grained ensemble docking search of the entire protein surface with two stages of atomistic flexible refinement. pepATTRACT yields high-quality predictions for 70 % of the cases when tested on a large benchmark of peptide-protein complexes. This performance in fully blind mode is similar to state-of-the-art local docking approaches that use information on the location of the binding site. Limiting the search to the peptide-binding region, the resulting pepATTRACT-local approach further improves the performance. Docking scripts for pepATTRACT and pepATTRACT-local can be generated via a web interface at www.attract.ph.tum.de/peptide.html . Here, we explain how to set up a docking run with the pepATTRACT web interface and demonstrate its usage by an application on binding of disordered regions from tumor suppressor p53 to a partner protein.

AnchorDock for Blind Flexible Docking of Peptides to Proteins

Due to increasing interest in peptides as signaling modulators and drug candidates, several methods for peptide docking to their target proteins are under active development. The "blind" docking problem, where the peptide-binding site on the protein surface is unknown, presents one of the current challenges in the field. AnchorDock protocol was developed by Ben-Shimon and Niv to address this challenge.This protocol narrows the docking search to the most relevant parts of the conformational space. This is achieved by pre-folding the free peptide and by computationally detecting anchoring spots on the surface of the unbound protein. Multiple flexible simulated annealing molecular dynamics (SAMD) simulations are subsequently carried out, starting from pre-folded peptide conformations, constrained to the various precomputed anchoring spots.Here, AnchorDock is demonstrated using two known protein-peptide complexes. A PDZ-peptide complex provides a relatively easy case due to the relatively small size of the protein, and a typical peptide conformation and binding region; a more challenging example is a complex between USP7_N-term and a p53-derived peptide, where the protein is larger, and the peptide conformation and a binding site are generally assumed to be unknown. AnchorDock returned native-like solutions ranked first and third for the PDZ and USP7 complexes, respectively. We describe the procedure step by step and discuss possible modifications where applicable.

Clustering of disulfide-rich peptides provides scaffolds for hit discovery by phage display: application to interleukin-23

BACKGROUND

Disulfide-rich peptides (DRPs) are found throughout nature. They are suitable scaffolds for drug development due to their small cores, whose disulfide bonds impart extraordinary chemical and biological stability. A challenge in developing a DRP therapeutic is to engineer binding to a specific target. This challenge can be overcome by (i) sampling the large sequence space of a given scaffold through a phage display library and by (ii) panning multiple libraries encoding structurally distinct scaffolds. Here, we implement a protocol for defining these diverse scaffolds, based on clustering structurally defined DRPs according to their conformational similarity.

RESULTS

We developed and applied a hierarchical clustering protocol based on DRP structural similarity, followed by two post-processing steps, to classify 806 unique DRP structures into 81 clusters. The 20 most populated clusters comprised 85% of all DRPs. Representative scaffolds were selected from each of these clusters; the representatives were structurally distinct from one another, but similar to other DRPs in their respective clusters. To demonstrate the utility of the clusters, phage libraries were constructed for three of the representative scaffolds and panned against interleukin-23. One library produced a peptide that bound to this target with an IC₅₀ of 3.3 μM.

CONCLUSIONS

Most DRP clusters contained members that were diverse in sequence, host organism, and interacting proteins, indicating that cluster members were functionally diverse despite having similar structure. Only 20 peptide scaffolds accounted for most of the natural DRP structural diversity, providing suitable starting points for seeding phage display experiments. Through selection of the scaffold surface to vary in phage display, libraries can be designed that present sequence diversity in architecturally distinct, biologically relevant combinations of secondary structures. We supported this hypothesis with a proof-of-concept experiment in which three phage libraries were constructed and panned against the IL-23 target, resulting in a single-digit μM hit and suggesting that a collection of libraries based on the full set of 20 scaffolds increases the potential to identify efficiently peptide binders to a protein target in a drug discovery program.

Analysis of the Yeast Peptidome and Comparison with the Human Peptidome

Peptides function as signaling molecules in species as diverse as humans and yeast. Mass spectrometry-based peptidomics techniques provide a relatively unbiased method to assess the peptidome of biological samples. In the present study, we used a quantitative peptidomic technique to characterize the peptidome of the yeast Saccharomyces cerevisiae and compare it to the peptidomes of mammalian cell lines and tissues. Altogether, 297 yeast peptides derived from 75 proteins were identified. The yeast peptides are similar to those of the human peptidome in average size and amino acid composition. Inhibition of proteasome activity with either bortezomib or epoxomicin led to decreased levels of some yeast peptides, suggesting that these peptides are generated by the proteasome. Approximately 30% of the yeast peptides correspond to the N- or C-terminus of the protein; the human peptidome is also highly represented in N- or C-terminal protein fragments. Most yeast and humans peptides are derived from a subset of abundant proteins, many with functions involving cellular metabolism or protein synthesis and folding. Of the 75 yeast proteins that give rise to peptides, 24 have orthologs that give rise to human and/or mouse peptides and for some, the same region of the proteins are found in the human, mouse, and yeast peptidomes. Taken together, these results support the hypothesis that intracellular peptides may have specific and conserved biological functions.

Integrative systems and synthetic biology of cell-matrix adhesion sites

The complexity of cell-matrix adhesion convolves its roles in the development and functioning of multicellular organisms and their evolutionary tinkering. Cell-matrix adhesion is mediated by sites along the plasma membrane that anchor the actin cytoskeleton to the matrix via a large number of proteins, collectively called the integrin adhesome. Fundamental challenges for understanding how cell-matrix adhesion sites assemble and function arise from their multi-functionality, rapid dynamics, large number of components and molecular diversity. Systems biology faces these challenges in its strive to understand how the integrin adhesome gives rise to functional adhesion sites. Synthetic biology enables engineering intracellular modules and circuits with properties of interest. In this review I discuss some of the fundamental questions in systems biology of cell-matrix adhesion and how synthetic biology can help addressing them.

Computational Studies on Structural, Excitation, and Charge-Transfer Properties of Ureidopeptidomimetics

Peptides with ureido group enclosing backbones are considered peptidomimetics and are known for their higher stabilities, biocompatibilities, antibiotic, inhibitor, and charge-transduction activities. These peptidomimetics have some unique applications, which are quite different from those of natural peptides. Hence, it is imperative to appreciate their properties at a microscopic level. In this regard, this work outlines, in detail, the charge transfer (CT) properties, hole-migration dynamics, and electronic structures of various experimentally comprehended ureidopeptidomimetic models using density functional theory (DFT). Time-dependent DFT and complete active space self-consistent field computations on basic models provide the necessary evidence for the viability of CT from the end enfolding the ureido group to the other end with a carboxylate entity. This donor-to-acceptor CT has been reflected in excitation studies, in which the higher intensity band corresponds to CT from the π orbital of the ureido group to the π* orbital of the carboxylate entity. Further, hole-migration studies have shown that charge can evolve from the ureido end, whereas the hole generated at the carboxylate end does not migrate. However, hole migration has been reported to occur from both ends (amino and carboxylate ends) in glycine oligopeptides, and our studies show that the ability to transfer and migrate charge can be tuned by modifying the donor and acceptor functional groups in both the neutral and cationic charge states. We have analyzed the possibility of hole migration following ionization using DFT-based wave-packet propagation and found its occurrence on a ∼2-5 fs time scale, which reflects the charge-transduction ability of peptidomimetics.

Molecular modelling, synthesis and biological evaluation of peptide inhibitors as anti-angiogenic agent targeting neuropilin-1 for anticancer application

Vascular endothelial growth factor (VEGF) and its co-receptor neuropilin-1 (NRP-1) are important targets of many pro-angiogenic factors. In this study, nine peptides were synthesized and evaluated for their molecular interaction with NRP-1 and compared to our previous peptide ATWLPPR. Docking study showed that the investigated peptides shared the same binding region as shown by tuftsin known to bind selectively to NRP-1. Four pentapeptides (DKPPR, DKPRR, TKPPR and TKPRR) and a hexapeptide CDKPRR demonstrated good inhibitory activity against NRP-1. In contrast, peptides having arginine residue at sites other than the C-terminus exhibited low activity towards NRP-1 and this is confirmed by their inability to displace the VEGF₁₆₅ binding to NRP-1. Docking study also revealed that replacement of carboxyl to amide group at the C-terminal arginine of the peptide did not affect significantly the binding interaction to NRP-1. However, the molecular affinity study showed that these peptides have marked reduction in the activity against NRP-1. Pentapeptides having C-terminal arginine showed strong interaction and good inhibitory activity with NRP thus may be a good template for anti-angiogenic targeting agent.

Reduced Levels of Proteasome Products in a Mouse Striatal Cell Model of Huntington's Disease

Huntington's disease is the result of a long polyglutamine tract in the gene encoding huntingtin protein, which in turn causes a large number of cellular changes and ultimately results in neurodegeneration of striatal neurons. Although many theories have been proposed, the precise mechanism by which the polyglutamine expansion causes cellular changes is not certain. Some evidence supports the hypothesis that the long polyglutamine tract inhibits the proteasome, a multiprotein complex involved in protein degradation. However, other studies report normal proteasome function in cells expressing long polyglutamine tracts. The controversy may be due to the methods used to examine proteasome activity in each of the previous studies. In the present study, we measured proteasome function by examining levels of endogenous peptides that are products of proteasome cleavage. Peptide levels were compared among mouse striatal cell lines expressing either 7 glutamines (STHdhQ7/Q7) or 111 glutamines in the huntingtin protein, either heterozygous (STHdhQ7/Q111) or homozygous (STHdhQ111/Q111). Both of the cell lines expressing huntingtin with 111 glutamines showed a large reduction in nearly all of the peptides detected in the cells, relative to levels of these peptides in cells homozygous for 7 glutamines. Treatment of STHdhQ7/Q7 cells with proteasome inhibitors epoxomicin or bortezomib also caused a large reduction in most of these peptides, suggesting that they are products of proteasome-mediated cleavage of cellular proteins. Taken together, these results support the hypothesis that proteasome function is impaired by the expression of huntingtin protein containing long polyglutamine tracts.

Modulating protein-protein interactions: the potential of peptides

Protein-protein interactions (PPIs) have emerged as important and challenging targets in chemical biology and medicinal chemistry. The main difficulty encountered in the discovery of small molecule modulators derives from the large contact surfaces involved in PPIs when compared with those that participate in protein-small molecule interactions. Because of their intrinsic features, peptides can explore larger surfaces and therefore represent a useful alternative to modulate PPIs. The use of peptides as therapeutics has been held back by their instability in vivo and poor cell internalization. However, more than 200 peptide drugs and homologous compounds (proteins or antibodies) containing peptide bonds are (or have been) on the market, and many alternatives are now available to tackle these limitations. This review will focus on the latest progress in the field, spanning from "lead" identification methods to binding evaluation techniques, through an update of the most successful examples described in the literature.

Origin of Helical Screw Sense Selectivity Induced by Chiral Constrained Cα-Tetrasubstituted α-Amino Acids in Aib-based Peptides

The mechanisms behind the propensity of chiral constrained Cα-tetrasubstituted amino acids (cCTAAs) to induce one particular helical screw sense, when included in the Ac-Aib2-cCTAA-Aib2-NHMe peptide model, were studied through replica exchange molecular dynamics, potential of mean force, and quantum theory of atoms in molecules calculations. We observed that cCTAAs exert their effect on helical screw sense selectivity through the positioning of the side chain to generate steric hindrance in either the (-x, +y, +z) or (+x, +y, -z) sectors of a right-handed 3D Cartesian space, where the z axis corresponds to the axis of the helix and the Cα lies on the +y semiaxis (0, +y, 0). The different strengthening of the noncovalent interactions, also comprising C-H···O interactions, exerted by the cCTAA in the right-handed or left-handed helix was also found important to define the preference of a cCTAA for a particular helix screw sense.

Fully Blind Peptide-Protein Docking with pepATTRACT

Peptide-protein interactions are ubiquitous in the cell and form an important part of the interactome. Computational docking methods can complement experimental characterization of these complexes, but current protocols are not applicable on the proteome scale. Here, we present a new fully blind flexible peptide-protein docking protocol, pepATTRACT, which combines a rapid coarse-grained global peptide docking search of the entire protein surface with a two-stage atomistic flexible refinement. Global unbound-unbound docking yielded near-native models for 70% of the docking cases when testing the protocol on the largest benchmark of peptide-protein complexes available to date. This performance is similar to that of state-of-the-art local docking protocols that rely on information about the binding site. Upon restricting the search to the peptide binding region, the resulting pepATTRACT-local approach outperformed existing methods. Docking scripts for pepATTRACT and pepATTRACT-local can be generated via a web interface at www.attract.ph.tum.de/peptide.html.

CABS-dock web server for the flexible docking of peptides to proteins without prior knowledge of the binding site

Protein–peptide interactions play a key role in cell functions. Their structural characterization, though challenging, is important for the discovery of new drugs. The CABS-dock web server provides an interface for modeling protein–peptide interactions using a highly efficient protocol for the flexible docking of peptides to proteins. While other docking algorithms require pre-defined localization of the binding site, CABS-dock does not require such knowledge. Given a protein receptor structure and a peptide sequence (and starting from random conformations and positions of the peptide), CABS-dock performs simulation search for the binding site allowing for full flexibility of the peptide and small fluctuations of the receptor backbone. This protocol was extensively tested over the largest dataset of non-redundant protein–peptide interactions available to date (including bound and unbound docking cases). For over 80% of bound and unbound dataset cases, we obtained models with high or medium accuracy (sufficient for practical applications). Additionally, as optional features, CABS-dock can exclude user-selected binding modes from docking search or to increase the level of flexibility for chosen receptor fragments. CABS-dock is freely available as a web server at http://biocomp.chem.uw.edu.pl/CABSdock.

AnchorDock: Blind and Flexible Anchor-Driven Peptide Docking

The huge conformational space stemming from the inherent flexibility of peptides is among the main obstacles to successful and efficient computational modeling of protein-peptide interactions. Current peptide docking methods typically overcome this challenge using prior knowledge from the structure of the complex. Here we introduce AnchorDock, a peptide docking approach, which automatically targets the docking search to the most relevant parts of the conformational space. This is done by precomputing the free peptide's structure and by computationally identifying anchoring spots on the protein surface. Next, a free peptide conformation undergoes anchor-driven simulated annealing molecular dynamics simulations around the predicted anchoring spots. In the challenging task of a completely blind docking test, AnchorDock produced exceptionally good results (backbone root-mean-square deviation ≤ 2.2Å, rank ≤15) for 10 of 13 unbound cases tested. The impressive performance of AnchorDock supports a molecular recognition pathway that is driven via pre-existing local structural elements.

Mechanism of stabilization of helix secondary structure by constrained Cα-tetrasubstituted α-amino acids

The theoretical basis behind the ability of constrained Cα-tetrasubstituted amino acids (CTAAs) to induce stable helical conformations has been studied through Replica Exchange Molecular Dynamics Potential of Mean Force Quantum Theory of Atoms In Molecules calculations on Ac-l-Ala-CTAA-l-Ala-Aib-l-Ala-NHMe peptide models. We found that the origin of helix stabilization by CTAAs can be ascribed to at least two complementary mechanisms limiting the backbone conformational freedom: steric hindrance predominantly in the (+x,+y,-z) sector of a right-handed 3D Cartesian space, where the z axis coincides with the helical axis and the Cα of the CTAA lies on the +y axis (0,+y,0), and the establishment of additional and relatively strong C-H···O interactions involving the CTAA.

Model peptides containing the 3-sulfanyl-norbornene amino acid, a conformationally constrained cysteine analogue effective inducer of 310-helix secondary structures

Two model peptides containing the 3-benzylsulfanylnorbornene amino acid (NRB) was prepared. Theoretical calculations, spectroscopic and X-ray analyses confirmed that both NRB enantiomers possess a strong right-handed helicogenic effect.

Knowledge-based modeling of peptides at protein interfaces: PiPreD

MOTIVATION

Protein-protein interactions (PPIs) underpin virtually all cellular processes both in health and disease. Modulating the interaction between proteins by means of small (chemical) agents is therefore a promising route for future novel therapeutic interventions. In this context, peptides are gaining momentum as emerging agents for the modulation of PPIs.

RESULTS

We reported a novel computational, structure and knowledge-based approach to model orthosteric peptides to target PPIs: PiPreD. PiPreD relies on a precompiled and bespoken library of structural motifs, iMotifs, extracted from protein complexes and a fast structural modeling algorithm driven by the location of native chemical groups on the interface of the protein target named anchor residues. PiPreD comprehensive and systematically samples the entire interface deriving peptide conformations best suited for the given region on the protein interface. PiPreD complements the existing technologies and provides new solutions for the disruption of selected interactions.

AVAILABILITY AND IMPLEMENTATION

Database and accessory scripts and programs are available upon request to the authors or at http://www.bioinsilico.org/PIPRED.

CONTACT

narcis.fernandez@gmail.com.

Designing of peptides with desired half-life in intestine-like environment

BACKGROUND

In past, a number of peptides have been reported to possess highly diverse properties ranging from cell penetrating, tumor homing, anticancer, anti-hypertensive, antiviral to antimicrobials. Owing to their excellent specificity, low-toxicity, rich chemical diversity and availability from natural sources, FDA has successfully approved a number of peptide-based drugs and several are in various stages of drug development. Though peptides are proven good drug candidates, their usage is still hindered mainly because of their high susceptibility towards proteases degradation. We have developed an in silico method to predict the half-life of peptides in intestine-like environment and to design better peptides having optimized physicochemical properties and half-life.

RESULTS

In this study, we have used 10mer (HL10) and 16mer (HL16) peptides dataset to develop prediction models for peptide half-life in intestine-like environment. First, SVM based models were developed on HL10 dataset which achieved maximum correlation R/R2 of 0.57/0.32, 0.68/0.46, and 0.69/0.47 using amino acid, dipeptide and tripeptide composition, respectively. Secondly, models developed on HL16 dataset showed maximum R/R2 of 0.91/0.82, 0.90/0.39, and 0.90/0.31 using amino acid, dipeptide and tripeptide composition, respectively. Furthermore, models that were developed on selected features, achieved a correlation (R) of 0.70 and 0.98 on HL10 and HL16 dataset, respectively. Preliminary analysis suggests the role of charged residue and amino acid size in peptide half-life/stability. Based on above models, we have developed a web server named HLP (Half Life Prediction), for predicting and designing peptides with desired half-life. The web server provides three facilities; i) half-life prediction, ii) physicochemical properties calculation and iii) designing mutant peptides.

CONCLUSION

In summary, this study describes a web server 'HLP' that has been developed for assisting scientific community for predicting intestinal half-life of peptides and to design mutant peptides with better half-life and physicochemical properties. HLP models were trained using a dataset of peptides whose half-lives have been determined experimentally in crude intestinal proteases preparation. Thus, HLP server will help in designing peptides possessing the potential to be administered via oral route (http://www.imtech.res.in/raghava/hlp/).

Proteasome inhibitors alter levels of intracellular peptides in HEK293T and SH-SY5Y cells

The proteasome cleaves intracellular proteins into peptides. Earlier studies found that treatment of human embryonic kidney 293T (HEK293T) cells with epoxomicin (an irreversible proteasome inhibitor) generally caused a decrease in levels of intracellular peptides. However, bortezomib (an antitumor drug and proteasome inhibitor) caused an unexpected increase in the levels of most intracellular peptides in HEK293T and SH-SY5Y cells. To address this apparent paradox, quantitative peptidomics was used to study the effect of a variety of other proteasome inhibitors on peptide levels in HEK293T and SH-SY5Y cells. Inhibitors tested included carfilzomib, MG132, MG262, MLN2238, AM114, and clasto-Lactacystin β-lactone. Only MG262 caused a substantial elevation in peptide levels that was comparable to the effect of bortezomib, although carfilzomib and MLN2238 elevated the levels of some peptides. To explore off-target effects, the proteosome inhibitors were tested with various cellular peptidases. Bortezomib did not inhibit tripeptidyl peptidase 2 and only weakly inhibited cellular aminopeptidase activity, as did some of the other proteasome inhibitors. However, potent inhibitors of tripeptidyl peptidase 2 (butabindide) and cellular aminopeptidases (bestatin) did not substantially alter the peptidome, indicating that the increase in peptide levels due to proteasome inhibitors is not a result of peptidase inhibition. Although we cannot exclude other possibilities, we presume that the paradoxical increase in peptide levels upon treatment with bortezomib and other inhibitors is the result of allosteric effects of these compounds on the proteasome. Because intracellular peptides are likely to be functional, it is possible that some of the physiologic effects of bortezomib and carfilzomib arise from the perturbation of peptide levels inside the cell.

PaFlexPepDock: parallel ab-initio docking of peptides onto their receptors with full flexibility based on Rosetta

Structural information related to protein–peptide complexes can be very useful for novel drug discovery and design. The computational docking of protein and peptide can supplement the structural information available on protein–peptide interactions explored by experimental ways. Protein–peptide docking of this paper can be described as three processes that occur in parallel: ab-initio peptide folding, peptide docking with its receptor, and refinement of some flexible areas of the receptor as the peptide is approaching. Several existing methods have been used to sample the degrees of freedom in the three processes, which are usually triggered in an organized sequential scheme. In this paper, we proposed a parallel approach that combines all the three processes during the docking of a folding peptide with a flexible receptor. This approach mimics the actual protein–peptide docking process in parallel way, and is expected to deliver better performance than sequential approaches. We used 22 unbound protein–peptide docking examples to evaluate our method. Our analysis of the results showed that the explicit refinement of the flexible areas of the receptor facilitated more accurate modeling of the interfaces of the complexes, while combining all of the moves in parallel helped the constructing of energy funnels for predictions.

AMPK-derived peptides reduce blood glucose levels but lead to fat retention in the liver of obese mice

AMP-activated protein kinase (AMPK) is a regulator of energy balance at both the cellular and the whole-body levels. Direct activation of AMPK has been highlighted as a potential novel, and possibly safer, alternative to treat type II diabetes and obesity. In this study, we aimed to design and characterize novel peptides that mimic the αG region of the α2 AMPK catalytic domain to modulate its activity by inhibiting interactions between AMPK domains or other interacting proteins. The derived peptides were tested in vivo and in tissue culture. The computationally predicted structure of the free peptide with the addition of the myristoyl (Myr) or acetyl (Ac) moiety closely resembled the protein structure that it was designed to mimic. Myr-peptide and Ac-peptide activated AMPK in muscle cells and led to reduced adipose tissue weight, body weight, blood glucose levels, insulin levels, and insulin resistance index, as expected from AMPK activation. In addition, triglyceride, cholesterol, leptin, and adiponectin levels were also lower, suggesting increased adipose tissue breakdown, a result of AMPK activation. On the other hand, liver weight and liver lipid content increased due to fat retention. We could not find an elevated pAMPK:AMPK ratio in the liver in vivo or in hepatocytes ex vivo, suggesting that the peptide does not lead to AMPK activation in hepatocytes. The finding that an AMPK-derived peptide leads to the activation of AMPK in muscle cells and in adipose tissue and leads to reduced glucose levels in obese mice, but to fat accumulation in the liver, demonstrates the differential effect of AMPK modulation in various tissues.

Druggable protein-protein interactions--from hot spots to hot segments

Protein-Protein Interactions (PPIs) mediate numerous biological functions. As such, the inhibition of specific PPIs has tremendous therapeutic value. The notion that these interactions are 'undruggable' has petered out with the emergence of more and more successful examples of PPI inhibitors, expanding considerably the scope of potential drug targets. The accumulated data on successes in the inhibition of PPIs allow us to analyze the features that are required for such inhibition. Whereas it has been suggested and shown that targeting hot spots at PPI interfaces is a good strategy to achieve inhibition, in this review we focus on the notion that the most amenable interactions for inhibition are those that are mediated by a 'hot segment', a continuous epitope that contributes the majority of the binding energy. This criterion is both useful in guiding future target selection efforts, and in suggesting immediate inhibitory candidates--the dominant peptidic segment that mediates the targeted interaction.

Peptide docking and structure-based characterization of peptide binding: from knowledge to know-how

Peptide-mediated interactions are gaining increased attention due to their predominant roles in the many regulatory processes that involve dynamic interactions between proteins. The structures of such interactions provide an excellent starting point for their characterization and manipulation, and can provide leads for targeted inhibitor design. The relatively few experimentally determined structures of peptide-protein complexes can be complemented with an outburst of modeling approaches that have been introduced in recent years, with increasing accuracy and applicability to ever more systems. We review different methods to address the considerable challenges in modeling the binding of a short yet highly flexible peptide to its partner. These methods apply an array of sampling strategies and draw from a recent amassing of knowledge about the biophysical nature of peptide-protein interactions. We elaborate on applications of these structure-based approaches and in particular on the characterization of peptide binding specificity to different peptide-binding domains and enzymes. Such applications can identify new biological targets and thus complement our current view of protein-protein interactions in living organisms. Accurate peptide-protein docking is of particular importance in the light of increased appreciation of the crucial functional roles of disordered regions and the many linear binding motifs embedded within.

Targeting protein-protein interactions as an anticancer strategy

The emergence and convergence of cancer genomics, targeted therapies, and network oncology have significantly expanded the landscape of protein-protein interaction (PPI) networks in cancer for therapeutic discovery. Extensive biological and clinical investigations have led to the identification of protein interaction hubs and nodes that are critical for the acquisition and maintenance of characteristics of cancer essential for cell transformation. Such cancer-enabling PPIs have become promising therapeutic targets. With technological advances in PPI modulator discovery and validation of PPI-targeting agents in clinical settings, targeting of PPI interfaces as an anticancer strategy has become a reality. Future research directed at genomics-based PPI target discovery, PPI interface characterization, PPI-focused chemical library design, and patient-genomic subpopulation-driven clinical studies is expected to accelerate the development of the next generation of PPI-based anticancer agents for personalized precision medicine. Here we briefly review prominent PPIs that mediate cancer-acquired properties, highlight recognized challenges and promising clinical results in targeting PPIs, and outline emerging opportunities.

A unified conformational selection and induced fit approach to protein-peptide docking

Protein-peptide interactions are vital for the cell. They mediate, inhibit or serve as structural components in nearly 40% of all macromolecular interactions, and are often associated with diseases, making them interesting leads for protein drug design. In recent years, large-scale technologies have enabled exhaustive studies on the peptide recognition preferences for a number of peptide-binding domain families. Yet, the paucity of data regarding their molecular binding mechanisms together with their inherent flexibility makes the structural prediction of protein-peptide interactions very challenging. This leaves flexible docking as one of the few amenable computational techniques to model these complexes. We present here an ensemble, flexible protein-peptide docking protocol that combines conformational selection and induced fit mechanisms. Starting from an ensemble of three peptide conformations (extended, a-helix, polyproline-II), flexible docking with HADDOCK generates 79.4% of high quality models for bound/unbound and 69.4% for unbound/unbound docking when tested against the largest protein-peptide complexes benchmark dataset available to date. Conformational selection at the rigid-body docking stage successfully recovers the most relevant conformation for a given protein-peptide complex and the subsequent flexible refinement further improves the interface by up to 4.5 Å interface RMSD. Cluster-based scoring of the models results in a selection of near-native solutions in the top three for ∼75% of the successfully predicted cases. This unified conformational selection and induced fit approach to protein-peptide docking should open the route to the modeling of challenging systems such as disorder-order transitions taking place upon binding, significantly expanding the applicability limit of biomolecular interaction modeling by docking.

Analysis of peptides secreted from cultured mouse brain tissue

Peptides represent a major class of cell-cell signaling molecules. Most peptidomic studies have focused on peptides present in brain or other tissues. For a peptide to function in intercellular signaling, it must be secreted. The present study was undertaken to identify the major peptides secreted from mouse brain slices that were cultured in oxygenated buffer for 3-4h. Approximately 75% of the peptides identified in extracts of cultured slices matched the previously reported peptide content of heat-inactivated mouse brain tissue, whereas only 2% matched the peptide content of unheated brain tissue; the latter showed a large number of postmortem changes. As found with extracts of heat-inactivated mouse brain, the extracts of cultured brain slices represented secretory pathway peptides as well as peptides derived from intracellular proteins such as those present in the cytosol and mitochondria. A subset of the peptides detected in the extracts of the cultured slices was detected in the culture media. The vast majority of secreted peptides arose from intracellular proteins and not secretory pathway proteins. The peptide RVD-hemopressin, a CB1 cannabinoid receptor agonist, was detected in culture media, which is consistent with a role for RVD-hemopressin as a non-classical neuropeptide. Taken together with previous studies, the present results show that short-term culture of mouse brain slices is an appropriate system to study peptide secretion, especially the non-conventional pathway(s) by which peptides produced from intracellular proteins are secreted. This article is part of a Special Issue entitled: An Updated Secretome.

Alterations of the intracellular peptidome in response to the proteasome inhibitor bortezomib

Bortezomib is an antitumor drug that competitively inhibits proteasome beta-1 and beta-5 subunits. While the impact of bortezomib on protein stability is known, the effect of this drug on intracellular peptides has not been previously explored. A quantitative peptidomics technique was used to examine the effect of treating human embryonic kidney 293T (HEK293T) cells with 5-500 nM bortezomib for various lengths of time (30 minutes to 16 hours), and human neuroblastoma SH-SY5Y cells with 500 nM bortezomib for 1 hour. Although bortezomib treatment decreased the levels of some intracellular peptides, the majority of peptides were increased by 50-500 nM bortezomib. Peptides requiring cleavage at acidic and hydrophobic sites, which involve beta-1 and -5 proteasome subunits, were among those elevated by bortezomib. In contrast, the proteasome inhibitor epoxomicin caused a decrease in the levels of many of these peptides. Although bortezomib can induce autophagy under certain conditions, the rapid bortezomib-mediated increase in peptide levels did not correlate with the induction of autophagy. Taken together, the present data indicate that bortezomib alters the balance of intracellular peptides, which may contribute to the biological effects of this drug.