Ignasi Belda

ENPDA: an evolutionary structure-based de novo peptide design algorithm

SummaryOne of the goals of computational chemists is to automate the de novo design of bioactive molecules. Despite significant advances in computational approaches to ligand design and binding energy evaluation, novel procedures for ligand design are required. Evolutionary computation provides a new approach to this design endeavor. We propose an evolutionary tool for de novo peptide design, based on the evaluation of energies for peptide binding to a user-defined protein surface patch. Special emphasis has been placed on the evaluation of the proposed peptides, leading to two different evaluation heuristics. The software developed was successfully tested on the design of ligands for the proteins prolyl oligopeptidase, p53, and DNA gyrase.

Mechanism of Binding of Fluoroquinolones to the Quinolone Resistance-Determining Region of DNA Gyrase: Towards an Understanding of the Molecular Basis of Quinolone Resistance

We have studied the bacterial resistance to fluoroquinolones that arises as a result of mutations in the DNA gyrase target protein. Although it is known that DNA gyrase is a target of quinolone antibacterial agents, the molecular details of the quinolone–gyrase interaction remain unclear. The mode of binding of ciprofloxacin, levofloxacin, and moxifloxacin to DNA gyrase was analyzed by means of docking calculations over the surface of the QRDR of GyrA. The analysis of these binding models allows study of the resistance mechanism associated with gyrA mutations more commonly found in E. coli fluoroquinolone‐resistant strains at the atomic level. Asp87 was found to be critical in the binding of these fluoroquinolones because it interacts with the positively charged nitrogens in these bactericidal drugs. The role of the other most common mutations at amino acid codon Ser83 can be explained through the contacts that the side chain of this residue establishes with fluoroquinolone molecules. Finally, our results strongly suggest that, although Arg121 has never been found to be associated with fluoroquinolone resistance, this residue makes a pivotal contribution to the binding of the antibiotic to GyrA and to defining its position in the QRDR of the enzyme.

Explicit Treatment of Water Molecules in Protein-Ligand Docking

No biological process can be fully described by computational techniques unless water is taken into consideration. Unfortunately, accurate representation of the water environment in any biological process is normally too timeconsuming with present techniques. Commonly, in ligand-binding docking, all water molecules are removed from the experimental structure data before the system is prepared, in which case it is assumed that all water effects are included in the scoring function. However, many scientific studies include particular water molecules explicitly, following chemical intuition. The choice of the number of water molecules and their positions depends substantially on the problem to be tackled and the information available about the problem. However, not much has been published comparing the quality of docking results with and without water molecules. This paper reviews those docking studies in which the effect of the inclusion of water molecules is analyzed in a large set of compounds. Explicit water molecules can be added to the system in different ways. The most popular option consists of including some water molecules already observed in crystal structures. However, tightly bound water molecules in a protein-ligand complex may not be conserved in complexes with other ligands. This problem can be tackled by coupling the prediction of water molecules with the docking calculation. Programs in which this feature is included are reviewed.

Computer-Aided Design of Fragment Mixtures for NMR-Based Screening

Fragment-based drug discovery is widely applied both in industrial and in academic screening programs. Several screening techniques rely on NMR to detect binding of a fragment to a target. NMR-based methods are among the most sensitive techniques and have the further advantage of yielding a low rate of false positives and negatives. However, NMR is intrinsically slower than other screening techniques; thus, to increase throughput in NMR-based screening, researchers often assay mixtures of fragments, rather than single fragments. Herein we present a fast and straightforward computer-aided method to design mixtures of fragments taken from a library that have minimized NMR signal overlap. This approach enables direct identification of one or several active fragments without the need for deconvolution. Our approach entails encoding of NMR spectra into a computer-readable format that we call a fingerprint, and minimizing the global signal overlap through a Monte Carlo algorithm. The scoring function used favors a homogenous distribution of the global signal overlap. The method does not require additional experimental work: the only data required are NMR spectra, which are generally recorded for each compound as a quality control measure before its insertion into the library.

Evolutionary algorithms and de novo peptide design

One of the goals of computational chemistry is the automated de novo design of bioactive molecules. Despite significant progress in computational approaches to ligand design and efficient evaluation of binding energy, novel procedures for ligand design are required. Evolutionary computation provides a new approach to this design issue. This paper presents an automated methodology for computer-aided peptide design based on evolutionary algorithms. It provides an automatic tool for peptide de novo design, based on protein surface patches defined by user. Regarding the restrictive constrains of this problem a special emphasis has been made on the design of the evolutionary algorithms implemented.

Design of enhanced agonists through the use of a new virtual screening method: Application to peptides that bind class I major histocompatibility complex (MHC) molecules

A new screening procedure is described that uses docking calculations to design enhanced agonist peptides that bind to major histocompatibility complex (MHC) class I receptors. The screening process proceeds via single mutations of one amino acid at the positions that directly interact with the MHC receptor. The energetic and structural effects of these mutations have been studied using fragments of the original ligand that vary in length. The results of these docking studies indicate that the mutant affinity ranking of long peptides can be practically reproduced with a screening approach performed using fragments of six residues. Fragments of four and five residues could mimic, in some cases, the structural arrangement of the side chains of the full‐length peptide. We have compared the structural and energetic results of the docking calculations with experimental data using three unrelated ligand peptides that differ greatly in their affinity for the MHC complex. Analysis of the affinity of the fragments led to the identification of three important parameters in the construction of fragments that mimic the structural and energetic properties of the full‐length ligand: the length of the fragment; its intermolecular energy; and the number and localization, internal or terminal, of the anchor residues. The results of this new peptide‐design methodology have been applied to suggest new peptides derived from the MUC1–8 peptide that could be used as murine vaccines that trigger the immune response through the MHC class I protein H‐2Kb.

Evolutionary computation and multimodal search: A good combination to tackle molecular diversity in the field of peptide design

SummaryThe awesome degree of structural diversity accessible in peptide design has created a demand for computational resources that can evaluate a multitude of candidate structures. In our specific case, we translate the peptide design problem to an optimization problem, and use evolutionary computation (EC) in tandem with docking to carry out a combinatorial search. However, the use of EC in huge search spaces with different optima may pose certain drawbacks. For example, EC is prone to focus a search in the first good region found. This is a problem not only because of the undesirable and automatic rejection of potentially good search space regions, but also because the found solution may be extremely difficult to synthesize chemically or may even be a false docking positive. In order to avoid rejecting potentially good solutions and to maximize the molecular diversity of the search, we have implemented evolutionary multimodal search techniques, as well as the molecular diversity metric needed by the multimodal algorithms to measure differences between various regions of the search space.

Chemically synthesized peptide libraries as a new source of BBB shuttles. Use of mass spectrometry for peptide identification.

The blood–brain barrier (BBB) is a biological barrier that protects the brain from neurotoxic agents and regulates the influx and efflux of molecules required for its correct function. This stringent regulation hampers the passage of brain parenchyma‐targeting drugs across the BBB. BBB shuttles have been proposed as a way to overcome this hurdle because these peptides can not only cross the BBB but also carry molecules which would otherwise be unable to cross the barrier unaided. Here we developed a new high‐throughput screening methodology to identify new peptide BBB shuttles in a broadly unexplored chemical space. By introducing d‐amino acids, this approach screens only protease‐resistant peptides. This methodology combines combinatorial chemistry for peptide library synthesis, in vitro models mimicking the BBB for library evaluation and state‐of‐the‐art mass spectrometry techniques to identify those peptides able to cross the in vitro assays. BBB shuttle synthesis was performed by the mix‐and‐split technique to generate a library based on the following: Ac‐d‐Arg‐XXXXX‐NH2, where X were: d‐Ala (a), d‐Arg (r), d‐Ile (i), d‐Glu (e), d‐Ser (s), d‐Trp (w) or d‐Pro (p). The assays used comprised the in vitro cell‐based BBB assay (mimicking both active and passive transport) and the PAMPA (mimicking only passive diffusion). The identification of candidates was determined using a two‐step mass spectrometry approach combining LTQ‐Orbitrap and Q‐trap mass spectrometers. Identified sequences were postulated to cross the BBB models. We hypothesized that some sequences cross the BBB through passive diffusion mechanisms and others through other mechanisms, including paracellular flux and active transport. These results provide a new set of BBB shuttle peptide families. Furthermore, the methodology described is proposed as a consistent approach to search for protease‐resistant therapeutic peptides. Copyright

Computer-Aided Peptide Evolution for Virtual Drug Design

One of the goals of computational chemistry is the automated de novo design of bioactive molecules. Despite significant progress in computational approaches to ligand design and efficient evaluation of binding energy, novel procedures for ligand design are required. Evolutionary computation provides a new approach to this design issue. A reliable framework for obtaining ligands via evolutionary algorithms has been implemented. It provides an automatic tool for peptide de novo design, based on protein surface patches defined by user. A special emphasis has been given to the evaluation of the proposed peptides. Hence, we have devised two different evaluation heuristics to carry out this task. Then, we have tested the proposed framework in the design of ligands for the protein Prolyl oligopetidase, p53, and DNA Gyrase.

Peptide data mining: from virtual design to knowledge extraction

Ignasi Belda, Ivan Traus, Susana Gordo, Teresa Tarrago, Sergio Madurga, Xavier Llora, and Ernest Giralt Institut de Recerca Biomedica, Parc Cientific de Barcelona, E 08028 Barcelona, Spain. b Conducive Corp. NY 10004, USA. c Illinois Genetic Algorithms Laboratory, Department of General Engineering, University of Illinois at Urbana-Champaign. IL 61801, USA. {ibelda,sgordo,ttarrago,smadurga,egiralt}@pcb.ub.edu, itraus@conducivecorp.com, xllora@illigal.ge.uiuc.edu