Extraordinary metabolic stability of peptides containing α-aminoxy acids

The metabolic stability of peptides containing a mixed sequence of α-aminoxy acids and α-amino acids is significantly improved compared to peptides composed of only natural α-amino acids. The introduction of an α-aminoxy acid into peptide chain dramatically improves the stability of the amide bonds immediately before and after it. These peptides containing α-aminoxy acids represent excellent structural scaffold for the design of metabolically stable and biologically active peptides.     

An on-bead assay for the identification of non-natural peptides targeting the androgen receptor-cofactor interaction

An efficient and rapid on-bead screening method was established to identify non-natural peptides that target the Androgen Receptor-cofactor interaction. Binding of the Androgen Receptor ligand binding domain to peptide sequences displayed on beads in a One-Bead-One-Compound format could be screened using fluorescence microscopy. The method was applied to generate and screen both a focussed and a random peptide library. Resynthesis of the peptide hits allowed for the verification of the affinity of the selected peptides for the Androgen Receptor in a competitive fluorescence polarization assay. For both libraries strong Androgen Receptor binding peptides were found, both with non-natural and natural amino acids. The peptides identified with natural amino acids showed great similarity in terms of preferred amino acid sequence with peptides previously isolated from biological screens, thus validating the screening approach. The non-natural peptides featured important novel chemical transformations on the relevant hydrophobic amino acid positions interacting with the Androgen Receptor. This screening approach expands the molecular diversity of peptide inhibitors for nuclear receptors.     

Characterization of purified c-type heme-containing peptides and identification of c-type heme-attachment sites in Shewanella oneidenis cytochromes using mass spectrometry

We describe methods for mass spectrometric identification of heme-containing peptides from c-type cytochromes that contain the CXXCH (X=any amino acid) sequence motif. The heme fragment ion yielded the most abundant MS/MS peak for standard heme-containing peptides with one amino acid difference for both 2+ and 3+ peptide charge states; both sequence and charge affect the extent of heme loss. Application to Shewanella oneidenis demonstrated the utility of this approach for identifying c-type heme-containing peptides from complex proteome samples.     

Automatic Edman microsequencing of peptides containing multiple unnatural amino acids

It is now routine using automatic Edman microsequencing to determine the primary structure of peptides or proteins containing natural amino acids; however, a deficiency in the ability to readily sequence peptides containing unnatural amino acids remains. With the advent of synthetic peptide chemistry, combinatorial chemistry, and the large number of commercially available unnatural amino acids, there is a need for efficient and accurate structure determination of short peptides containing many unnatural amino acids. In this study, 35 commercially available alpha-unnatural amino acids were selected to determine their elution profile on an ABI protein sequencer. Using a slightly modified gradient program, 19 of these 35 PTH amino acids can be readily resolved and distinguished from common PTH amino acids at low picomole levels. These unnatural amino acids in conjunction with the 20 natural amino acids can be used as building blocks to construct peptide libraries, and peptide beads isolated from these libraries can be readily microsequenced. To demonstrate this, we synthesized a simple tripeptide "one-bead one-compound" combinatorial library containing 14 unnatural and 19 natural amino acids and screened this library for streptavidin-binding ligands. Microsequencing of the isolated peptide-beads revealed the novel motif Bpa-Phe(4-X)-Aib, wherein X = H, OH, and CH3.     

In silico identification of new secretory peptide genes in Drosophila melanogaster

Bioactive peptides play critical roles in regulating most biological processes in animals. The elucidation of the amino acid sequence of these regulatory peptides is crucial for our understanding of animal physiology. Most of the (neuro)peptides currently known were identified by purification and subsequent amino acid sequencing. With the entire genome sequence of some animals now available, it has become possible to predict novel putative peptides. In this way, BLAST (Basic Local Alignment Searching Tool) analysis of the Drosophila melanogaster genome has allowed annotation of 36 secretory peptide genes so far. Peptide precursor genes are, however, poorly predicted by this algorithm, thus prompting an alternative approach described here. With the described searching program we scanned the Drosophila genome for predicted proteins with the structural hallmarks of neuropeptide precursors. As a result, 76 additional putative secretory peptide genes were predicted in addition to the 43 annotated ones. These putative (neuro)peptide genes contain conserved motifs reminiscent of known neuropeptides from other animal species. Peptides that display sequence similarities to the mammalian vasopressin, atrial natriuretic peptide, and prolactin precursors and the invertebrate peptides orcokinin, prothoracicotropic hormones, trypsin modulating oostatic factor, and Drosophila immune induced peptides (DIMs) among others were discovered. Our data hence provide further evidence that many neuropeptide genes were already present in the ancestor of Protostomia and Deuterostomia prior to their divergence. This bioinformatic study opens perspectives for the genome-wide analysis of peptide genes in other eukaryotic model organisms.     

Extensive post-translational modification, including serine to D-alanine conversion, in the two-component lantibiotic, lacticin 3147

Lacticin 3147 is a two-component bacteriocin produced by Lactococcus lactis subspecies lactis DPC3147. In order to further characterize the biochemical nature of the bacteriocin, both peptides were isolated which together are responsible for the antimicrobial activity. The first, LtnA1, is a 3,322 Da 30-amino acid peptide and the second component, LtnA2, is a 29-amino acid peptide with a mass of 2,847 Da. Conventional amino acid analysis revealed that both peptides contain the thioether amino acid, lanthionine, as well as an excess of alanine to that predicted from the genetic sequence of the peptides. Chiral phase gas chromatography coupled with mass spectrometry of amino acid composition indicated that both LtnA1 and LtnA2 contain D-alanine residues and amino acid sequence analysis of LtnA1 confirmed that the D-alanine results from post-translational modification of a serine residue in the primary translation product. Taken together, these results demonstrate that lacticin 3147 is a novel, two-component, D-alanine containing lantibiotic that undergoes extensive post-translational modification which may account for its potent antimicrobial activity against a wide range of Gram-positive bacteria.     

Using a neural network to identify potential HLA-DR1 binding sites within proteins

The presentation by antigen-presenting cells of immunodominant peptide segments in association with major histocompatibility complex (MHC) encoded proteins is fundamental to the efficacy of a specific immune response. One approach used to identify immunodominant segments within proteins has involved the development of predictive algorithms which utilize amino acid sequence data to identify structural characteristics or motifs associated with in vivo antigenicity. The parallel-computing technique termed ‘neural networking’ has recently been shown to be remarkably efficient at addressing the problem of pattern recognition and can be applied to predict protein secondary structure attributes directly from amino acid sequence data. In order to examine the potential of a neural network to generalize peptide structural feature related to binding within class II MHC-encoded proteins, we have trained a neural network to determine whether or not any given amino acid of a protein is part of a peptide segment capable of binding to HLA-DR1. We report that a neural network trained on a data base consisting of peptide segments known to bind to HLA-DR1 is able to generalize features relating to HLA-DR1-binding capacity (r = 0.17 and p = 0.0001).     

AMS 3.0: prediction of post-translational modifications

Background  

We present here the recent update of AMS algorithm for identification of post-translational modification (PTM) sites in proteins based only on sequence information, using artificial neural network (ANN) method. The query protein sequence is dissected into overlapping short sequence segments. Ten different physicochemical features describe each amino acid; therefore nine residues long segment is represented as a point in a 90 dimensional space. The database of sequence segments with confirmed by experiments post-translational modification sites are used for training a set of ANNs.     

A novel knowledge-based approach to design inorganic-binding peptides

MOTIVATION: The discovery of solid-binding peptide sequences is accelerating along with their practical applications in biotechnology and materials sciences. A better understanding of the relationships between the peptide sequences and their binding affinities or specificities will enable further design of novel peptides with selected properties of interest both in engineering and medicine. RESULTS: A bioinformatics approach was developed to classify peptides selected by in vivo techniques according to their inorganic solid-binding properties. Our approach performs all-against-all comparisons of experimentally selected peptides with short amino acid sequences that were categorized for their binding affinity and scores the alignments using sequence similarity scoring matrices. We generated novel scoring matrices that optimize the similarities within the strong-binding peptide sequences and the differences between the strong- and weak-binding peptide sequences. Using the scoring matrices thus generated, a given peptide is classified based on the sequence similarity to a set of experimentally selected peptides. We demonstrate the new approach by classifying experimentally characterized quartz-binding peptides and computationally designing new sequences with specific affinities. Experimental verifications of binding of these computationally designed peptides confirm our predictions with high accuracy. We further show that our approach is a general one and can be used to design new sequences that bind to a given inorganic solid with predictable and enhanced affinity.     

Identification of a heparin binding domain of the neural cell adhesion molecule N-CAM using synthetic peptides

The neural cell adhesion molecule (N-CAM) plays an integral role in cell interactions during neural development, with the binding of heparan sulfate proteoglycan to the amino-terminal region of N-CAM being required for N-CAM function. In the present study we have used synthetic peptides (HBD-1 and HBD-2), derived from the primary amino acid sequence of rat N-CAM, to identify the region of N-CAM that binds heparan sulfate. The 28 amino acid HBD-1 synthetic peptide was shown to bind both [3H]heparin and dissociated retinal cells. Retinal cells also attach to a substratum of HBD-2 peptide, but fail to bind to a control peptide containing a scrambled amino acid sequence of HBD-2. The HBD-2 peptide also inhibits retinal cell adhesion to N-CAM, demonstrating the physiological importance of the amino acid sequence encoded by the HBD peptide. These data therefore permit the localization of a heparin binding domain to a 17 amino acid region of immunoglobulin-like loop 2.     

A unified multitask architecture for predicting local protein properties

A variety of functionally important protein properties, such as secondary structure, transmembrane topology and solvent accessibility, can be encoded as a labeling of amino acids. Indeed, the prediction of such properties from the primary amino acid sequence is one of the core projects of computational biology. Accordingly, a panoply of approaches have been developed for predicting such properties; however, most such approaches focus on solving a single task at a time. Motivated by recent, successful work in natural language processing, we propose to use multitask learning to train a single, joint model that exploits the dependencies among these various labeling tasks. We describe a deep neural network architecture that, given a protein sequence, outputs a host of predicted local properties, including secondary structure, solvent accessibility, transmembrane topology, signal peptides and DNA-binding residues. The network is trained jointly on all these tasks in a supervised fashion, augmented with a novel form of semi-supervised learning in which the model is trained to distinguish between local patterns from natural and synthetic protein sequences. The task-independent architecture of the network obviates the need for task-specific feature engineering. We demonstrate that, for all of the tasks that we considered, our approach leads to statistically significant improvements in performance, relative to a single task neural network approach, and that the resulting model achieves state-of-the-art performance.     

Cross-validation of protein structural class prediction using statistical clustering and neural networks.

We present an approach to predicting protein structural class that uses amino acid composition and hydrophobic pattern frequency information as input to two types of neural networks: (1) a three-layer back-propagation network and (2) a learning vector quantization network. The results of these methods are compared to those obtained from a modified Euclidean statistical clustering algorithm. The protein sequence data used to drive these algorithms consist of the normalized frequency of up to 20 amino acid types and six hydrophobic amino acid patterns. From these frequency values the structural class predictions for each protein (all-alpha, all-beta, or alpha-beta classes) are derived. Examples consisting of 64 previously classified proteins were randomly divided into multiple training (56 proteins) and test (8 proteins) sets. The best performing algorithm on the test sets was the learning vector quantization network using 17 inputs, obtaining a prediction accuracy of 80.2%. The Matthews correlation coefficients are statistically significant for all algorithms and all structural classes. The differences between algorithms are in general not statistically significant. These results show that information exists in protein primary sequences that is easily obtainable and useful for the prediction of protein structural class by neural networks as well as by standard statistical clustering algorithms.     

Definition of the M-conotoxin superfamily: characterization of novel peptides from molluscivorous Conus venoms

Most of the >50,000 different pharmacologically active peptides in Conus venoms belong to a small number of gene superfamilies. In this work, the M-conotoxin superfamily is defined using both biochemical and molecular criteria. Novel excitatory peptides purified from the venoms of the molluscivorous species Conus textile and Conus marmoreus all have a characteristic pattern of Cys residues previously found in the mu-, kappaM-, and psi-conotoxins (CC-C-C-CC). The new peptides are smaller (12-19 amino acids) than the mu-, kappaM-, and psi-conotoxins (22-24 amino acids). One peptide, mr3a, was chemically synthesized in a biologically active form. Analysis of the disulfide bridges of a natural peptide tx3c from C. textile and synthetic peptide mr3a from C. marmoreus showed a novel pattern of disulfide connectivity, different from that previously established for the mu- and psi-conotoxins. Thus, these peptides belong to a new group of structurally and pharmacologically distinct conotoxins that are particularly prominent in the venoms of mollusc-hunting Conus species. Analysis of cDNA clones encoding the novel peptides as well as those encoding mu-, kappaM-, and psi-conotoxins revealed highly conserved amino acid residues in the precursor sequences; this conservation in both amino acid sequence and in the Cys pattern defines a gene superfamily, designated the M-conotoxin superfamily. The peptides characterized can be provisionally assigned to four distinct groups within the M-superfamily based on sequence similarity within and divergence between each group. A notable feature of the superfamily is that two distinct structural frameworks have been generated by changing the disulfide connectivity on an otherwise conserved Cys pattern.     

Modeling an active conformation for linear peptides and design of a competitive inhibitor for HMG-CoA reductase

This study presents an approach that can be used to search for lead peptide candidates, including unconstrained structures in a recognized sequence. This approach was performed using the design of a competitive inhibitor for 3-hydroxy-3-methylglutaryl CoA reductase (HMGR). In a previous design for constrained peptides, a head-to-tail cyclic structure of peptide was used as a model of linear analog in searches for lead peptides with a structure close to an active conformation. Analysis of the conformational space occupied by the peptides suggests that an analogical approach can be applied for finding a lead peptide with an unconstrained structure in a recognized sequence via modeling a cycle using fixed residues of the peptide backbone. Using the space obtained by an analysis of the bioactive conformations of statins, eight cyclic peptides were selected for a peptide library based on the YVAE sequence as a recognized motif. For each cycle, the four models were assessed according to the design criterion ("V" parameter) applied for constrained peptides. Three cyclic peptides (FGYVAE, FPYVAE, and FFYVAE) were selected as lead cycles from the library. The linear FGYVAE peptide (IC(50) = 0.4 microM) showed a 1200-fold increase the inhibitory activity compared to the first isolated LPYP peptide (IC(50) = 484 microM) from soybean. Experimental analysis of the modeled peptide structures confirms the appropriateness of the proposed approach for the modeling of active conformations of peptides.     

De novo sequence analysis of N-terminal sulfonated peptides after in-gel guanidination

Here we report a novel approach in which gel-separated proteins are guanidinated in-gel prior to enzymatic cleavage. In contrast to previously described techniques, this procedure allows the extracted tryptic peptides to be N-terminal sulfonated without any further sample purification. The derivatized peptides were subsequently fragmented using a matrix-assisted laser desorption/ionization time of flight/time of flight instrument. The approach facilitates the de novo sequence analysis and allows obtaining longer stretches of amino acid sequence information. We demonstrate that the obtained information can be used to identify proteins using a sequence similarity search algorithm. The technique was compared to the standard peptide mass fingerprint approach, applied either in-gel or in solution, using a number of sodium dodecyl sulfate-polyacrylamide gel electrophoresis separated model proteins. Finally, the new protocol was applied on a proteomic study of two-dimensional PAGE separated proteins from Shewanella oneidensis. More than 50 proteins from this organism were identified using sub-picomol quantities of protein, and peptide sequences of up to 20 amino acid residues in length have been determined.     

Design of inhibitors of Ras--Raf interaction using a computational combinatorial algorithm

Drugs that inhibit important protein-protein interactions are hard to find either by screening or rational design, at least so far. Most drugs on the market that target proteins today are therefore aimed at well-defined binding pockets in proteins. While computer-aided design is widely used to facilitate the drug discovery process for binding pockets, its application to the design of inhibitors that target the protein surface initially seems to be limited because of the increased complexity of the task. Previously, we had started to develop a computational combinatorial design approach based on the well-known 'multiple copy simultaneous search' (MCSS) procedure to tackle this problem. In order to identify sequence patterns of potential inhibitor peptides, a three-step procedure is employed: first, using MCSS, the locations of specific functional groups on the protein surface are identified; second, after constructing the peptide main chain based on the location of favorite locations of N-methylacetamide groups, functional groups corresponding to amino acid side chains are selected and connected to the main chain C(alpha) atoms; finally, the peptides generated in the second step are aligned and probabilities of amino acids at each position are calculated from the alignment scheme. Sequence patterns of potential inhibitors are determined based on the propensities of amino acids at each C(alpha) position. Here we report the optimization of inhibitor peptides using the sequence patterns determined by our method. Several short peptides derived from our prediction inhibit the Ras--Raf association in vitro in ELISA competition assays, radioassays and biosensor-based assays, demonstrating the feasibility of our approach. Consequently, our method provides an important step towards the development of novel anti-Ras agents and the structure-based design of inhibitors of protein--protein interactions.     

A hybrid genetic-neural model for predicting protein structural classes

A genetic algorithm (GA) for feature selection in conjunction with neural network was applied to predict protein structural classes based on single amino acid and all dipeptide composition frequencies. These sequence parameters were encoded as input features for a GA in feature selection procedure and classified with a three-layered neural network to predict protein structural classes. The system was established through optimization of the classification performance of neural network which was used as evaluation function. In this study, self-consistency and jackknife tests on a database containing 498 proteins were used to verify the performance of this hybrid method, and were compared with some of prior works. The adoption of a hybrid model, which encompasses genetic and neural technologies, demonstrated to be a promising approach in the task of protein structural class prediction.