Publications, N M Glykos group

Word cloud of the abstracts that follow.

54. Vouzina, O.-D., Tafanidis, A. & Glykos*, N.M. (2024), "The curious case of A31P, a topology-switching mutant of the Repressor of Primer protein : A molecular dynamics study of its folding and misfolding", J. Chem. Inf. Model., 64, 6081-6091.
Electronic reprint (6.3 MBytes) : Local copy, or directly from J. Chem. Inf. Model.. See also the Supporting information file.

The effect of mutations on protein structures is usually rather localized and minor. Finding a mutation that can single-handedly change the fold and/or topology of a protein structure is a rare exception. The A31P mutant of the homodimeric Repressor of Primer (Rop) protein is one such exception: This single mutation -and as demonstrated by two independent crystal structure determinations- can convert the canonical (left-handed/all-antiparallel) 4-alpha-helical bundle of Rop, to a new form (right-handed/mixed parallel and antiparallel bundle) displaying a previously unobserved 'bisecting U' topology. The main problem with understanding the dramatic effect of this mutation on the folding of Rop is to understand its very existence : Most computational methods appear to agree that the mutation should have had no appreciable effect, with the majority of energy minimization methods and protein structure prediction protocols indicating that this mutation is fully consistent with the native Rop structure, requiring only a local and minor change at the mutation site. Here we use two long (10 us each) molecular dynamics simulations to compare the stability and dynamics of the native Rop versus a hypothetical structure that is identical with the native Rop but is carrying this single Alanine-31 to Proline mutation. Comparative analysis of the two trajectories convincingly shows that in contrast to the indications from energy minimization -but in agreement with the experimental data-, this hypothetical native-like A31P structure is unstable, with its turn regions almost completely unfolding, even under the relatively mild 320K NpT simulations that we have used for this study. We discuss the implication of these findings for the folding of the A31P mutant, especially with respect to the proposed model of a double-funneled energy landscape.

53. Kolocouris*, A., Arkin, I. & Glykos*, N.M. (2022), "A proof-of-concept study of the secondary structure of influenza A, B M2 and MERS- and SARS-CoV E transmembrane peptides using folding molecular dynamics simulations in a membrane mimetic solvent", Phys. Chem. Chem. Phys., 24, 25391-25402.
Electronic reprint (1.8 MBytes) : Local copy, or directly from Phys. Chem. Chem. Phys., Copyright © Royal Society of Chemistry.

We have carried out a proof-of-concept molecular dynamics (MD) simulation with adaptive tempering in a membrane mimetic environment to study the folding of single-pass membrane peptides. We tested the influenza A M2 viroporin, influenza B M2 viroporin, and protein E from coronaviruses MERS-Cov-2 and SARS-CoV-2 peptides with known experimental secondary structures in membrane bilayers. The two influenza-derived peptides are significantly different in the peptide sequence and secondary structure and more polar than the two coronavirus-derived peptides. Through a total of more than 50 μs of simulation time that could be accomplished in trifluoroethanol (TFE), as a membrane model, we characterized comparatively the folding behavior, helical stability, and helical propensity of these transmembrane peptides that match perfectly their experimental secondary structures, and we identified common motifs that reflect their quaternary organization and known (or not) biochemical function. We showed that BM2 is organized into two structurally distinct parts: a significantly more stable N-terminal half, and a fast-converting C-terminal half that continuously folds and unfolds between α-helical structures and non-canonical structures, which are mostly turns. In AM2, both the N-terminal half and C-terminal half are very flexible. In contrast, the two coronavirus-derived transmembrane peptides are much more stable and fast helix-formers when compared with the influenza ones. In particular, the SARS-derived peptide E appears to be the fastest and most stable helix-former of all the four viral peptides studied, with a helical structure that persists almost without disruption for the whole of its 10 μs simulation. By comparing the results with experimental observations, we benchmarked TFE in studying the conformation of membrane and hydrophobic peptides. This work provided accurate results suggesting a methodology to run long MD simulations and predict structural properties of biologically important membrane peptides.

52. Gkogka, I. & Glykos*, N.M. (2022), "Folding molecular dynamics simulation of T-peptide, a HIV viral entry inhibitor : Structure, dynamics, and comparison with the experimental data", J. Comput. Chem., 43, 942-952.
Electronic reprint (18 MBytes) : Local copy, or directly from J. Comput. Chem., Copyright © Wiley Periodicals, Inc.

Peptide T is a synthetic octapeptide fragment, which corresponds to the region 185–192 of the gp120 HIV coat protein and functions as a viral entry inhibitor. In this work, a folding molecular dynamics simulation of peptide T in a membrane-mimicking (DMSO) solution was performed with the aim of characterizing the peptide's structural and dynamical properties. We show that peptide T is highly flexible and dynamic. The main structural characteristics observed were rapidly interconverting short helical stretches and turns, with a notable preference for the formation of β-turns. The simulation also indicated that the C-terminal part appears to be more stable than the rest of the peptide, with the most preferred conformation for residues 5–8 being a β-turn. In order to validate the accuracy of the simulations, we compared our results with the experimental NMR data obtained for the T-peptide in the same solvent. In agreement with the simulation, the NMR data indicated the presence of a preferred structure in solution that was consistent with a β-turn comprising the four C-terminal residues. An additional comparison between the experimental and simulation-derived chemical shifts also showed a reasonable agreement between experiment and simulation, further validating the simulation-derived structural characterization of the T-peptide. We conclude that peptide folding simulations produce physically relevant results even when performed with organic solvents that were not part of the force field parameterization procedure.

51. Mitsikas, D.A. & Glykos*, N.M. (2020), "A molecular dynamics simulation study on the propensity of Asn-Gly-containing heptapeptides towards β-turn structures: Comparison with ab initio quantum mechanical calculations", PLoS ONE, 15(12): e0243429.
Electronic reprint (2.8 MBytes) : Local copy, or directly from PLoS ONE, Copyright © Mitsikas & Glykos under the Creative Commons Attribution License.

Both molecular mechanical and quantum mechanical calculations play an important role in describing the behavior and structure of molecules. In this work, we compare for the same peptide systems the results obtained from folding molecular dynamics simulations with previously reported results from quantum mechanical calculations. More specifically, three molecular dynamics simulations of 5 μs each in explicit water solvent were carried out for three Asn-Gly-containing heptapeptides, in order to study their folding and dynamics. Previous data, based on quantum mechanical calculations within the DFT framework have shown that these peptides adopt β-turn structures in aqueous solution, with type I’ β-turn being the most preferred motif. The results from our analyses indicate that at least for the given systems, force field and simulation protocol, the two methods diverge in their predictions. The possibility of a force field-dependent deficiency is examined as a possible source of the observed discrepancy.

50. Kokkinidis, M., Glykos*, N.M. & Fadouloglou*, V.E. (2020), "Catalytic activity regulation through post-translational modification: the expanding universe of protein diversity", Adv. Protein Chem. Struct. Biol., 122, 97-125.
Electronic reprint (2.5 MBytes) : Local copy, or directly from Adv. Protein Chem. Struct. Biol., Copyright © Elsevier.

Protein composition is restricted by the genetic code to a relatively small number of natural amino acids. Similarly, the known three-dimensional structures adopt a limited number of protein folds. However, proteins exert a large variety of functions and show a remarkable ability for regulation and immediate response to intracellular and extracellular stimuli. To some degree, the wide variability of protein function can be attributed to the post-translational modifications. Post-translational modifications have been observed in all kingdoms of life and give to proteins a significant degree of chemical and consequently functional and structural diversity. Their importance is partly reflected in the large number of genes dedicated to their regulation. So far, hundreds of post-translational modifications have been observed while it is believed that many more are to be discovered along with the technological advances in sequencing, proteomics, mass spectrometry and structural biology. Indeed, the number of studies which report novel post translational modifications is getting larger supporting the notion that their space is still largely unexplored. In this review we explore the impact of post-translational modifications on protein structure and function with emphasis on catalytic activity regulation. We present examples of proteins and protein families whose catalytic activity is substantially affected by the presence of post translational modifications and we describe the molecular basis which underlies the regulation of the protein function through these modifications. When available, we also summarize the current state of knowledge on the mechanisms which introduce these modifications to protein sites.

49. Stylianakis, I., Shalev, A., Scheiner, S., Sigalas, M.P., Arkin, I.T., Glykos*, N.M. & Kolocouris*, A. (2020), "The balance between side‐chain and backbone‐driven association in folding of the α‐helical influenza A transmembrane peptide", J. Comput. Chem., 41, 2177-2188.
Electronic reprint (3.1 MBytes) : Local copy, or directly from J. Comput. Chem., Copyright © Wiley Periodicals, Inc.

The correct balance between attractive, repulsive and peptide hydrogen bonding interactions must be attained for proteins to fold correctly. To investigate these important contributors, we sought a comparison of the folding between two 25‐residues peptides, the influenza A M2 protein transmembrane domain (M2TM) and the 25‐Ala (Ala25). M2TM forms a stable α‐helix as is shown by circular dichroism (CD) experiments. Molecular dynamics (MD) simulations with adaptive tempering show that M2TM monomer is more dynamic in nature and quickly interconverts between an ensemble of various α‐helical structures, and less frequently turns and coils, compared to one α‐helix for Ala25. DFT calculations suggest that folding from the extended structure to the α‐helical structure is favored for M2TM compared with Ala25. This is due to CH⋯O attractive interactions which favor folding to the M2TM α‐helix, and cannot be described accurately with a force field. Using natural bond orbital (NBO) analysis and quantum theory atoms in molecules (QTAIM) calculations, 26 CH⋯O interactions and 22 NH⋯O hydrogen bonds are calculated for M2TM. The calculations show that CH⋯O hydrogen bonds, although individually weaker, have a cumulative effect that cannot be ignored and may contribute as much as half of the total hydrogen bonding energy, when compared to NH⋯O, to the stabilization of the α‐helix in M2TM. Further, a strengthening of NH⋯O hydrogen bonding interactions is calculated for M2TM compared to Ala25. Additionally, these weak CH⋯O interactions can dissociate and associate easily leading to the ensemble of folded structures for M2TM observed in folding MD simulations.

48. Riziotis, I.G. & Glykos*, N.M. (2019), "On the presence of short-range periodicities in protein structures that are not related to established secondary structure elements", Proteins, 87, 966-978.
Electronic reprint (5.7 MBytes) : Local copy, or directly from Proteins, Copyright © Wiley. See also the Supporting information file (2.3 Mbytes).

Standard secondary structure elements such as α-helices or β-sheets, are characterized by repeating backbone torsion angles (φ,ψ) at the single residue level. Two-residue motifs of the type (φ,ψ)₂ are also observed in non-linear conformations, mainly turns. Taking these observations a step further, it can be argued that there is no a priori reason why the presence of higher order periodicities can not be envisioned in protein structures, such as, for example, periodic transitions between successive residues of the type (...-α-β-α-β-α-...), or, (...-β-αL-β-αL-β-...), or, (...-α-β-αL-α-β-αL-...), etc., where the symbols (α,β,αL) refer to the established Ramachandran-based residue conformations. From all such possible higher order periodicities, here we examine the deposited (with the PDB) protein structures for the presence of short-range periodical conformations comprising five consecutive residues alternating between two (and only two) distinct Ramachandran regions, for example conformations of the type (α-β-α-β-α) or (β-αL-β-αL-β) etc. Using a probabilistic approach we have located several thousand of such peptapeptides, and these were clustered and analyzed in terms of their structural characteristics, their sequences, and their putative functional correlations using a gene ontology-based approach. We show that such non-standard short-range periodicities are present in a large and functionally diverse sample of proteins, and can be grouped into two structurally conserved major types. Examination of the structural context in which these peptapeptides are observed gave no conclusive evidence for the presence of a persistent structural or functional role of these higher order periodic conformations.

47. Georgoulia, P.S. & Glykos*, N.M. (2019), "Molecular simulation of peptides coming of age: Accurate prediction of folding, dynamics and structures", Arch. Biochem. Biophys., 664, 76-88.
Electronic reprint (1.6 MBytes) : Local copy, or directly from Arch. Biochem. Biophys., Copyright © Elsevier Inc.

The application of molecular dynamics simulations to study the folding and dynamics of peptides has attracted a lot of interest in the last couple of decades. Following the successful prediction of the folding of several proteins using molecular simulation, foldable peptides emerged as a favourable system mainly due to their application in improving protein structure prediction methods and in drug design studies. However, their performance is inherently linked to the accuracy of the empirical force fields used in the simulations, whose optimisation and validation is of paramount importance. Here we review the most important findings in the field of molecular peptide simulations and highlight the significant advancements made over the last twenty years. Special reference is made on the simulation of disordered peptides and the remaining challenge to find a force field able to describe accurately their conformational landscape.

46. Georgoulia*, P.S. & Glykos, N.M. (2018), "Folding Molecular Dynamics Simulation of a gp41-Derived Peptide Reconcile Divergent Structure Determinations", ACS Omega, 3, 14746-14754.
Electronic reprint (1.7 MBytes) : Local copy, or directly from ACS Omega, Copyright © American Chemical Society. See also the Supporting information file (11 Mbytes).

T-20 peptide is the first FDA-approved fusion inhibitor against AIDS/HIV-1 gp41 protein. Part of it, the gp41[659-671] peptide, that contains the complete epitope for the neutralizing 2F5 monoclonal antibody, has been found experimentally in a number of divergent structures. Herein, we attempt to reconcile them by using unbiased large-scale all-atom molecular dynamics folding simulations. We show that our approach can successfully capture the peptide's heterogeneity and reach each and every experimentally determined conformation in sub-angstrom accuracy, whilst preserving the peptide's disordered nature. Our analysis also unveils that the minor refinements within the AMBER99SB family of force fields can lead to appreciable differences in the predicted conformational stability arising from subtle differences in the helical- and Î²-region of the Ramachandran plot. Our work underlines the contribution of molecular dynamics simulation in structurally characterizing pharmacologically important peptides of ambiguous structure.

45. Adamidou, T., Arvaniti, K.-O. & Glykos*, N.M. (2018), "Folding Simulations of a Nuclear Receptor Box-Containing Peptide Demonstrate the Structural Persistence of the LxxLL Motif Even in the Absence of Its Cognate Receptor", J. Phys. Chem. B, 122, 106â116.
Electronic reprint (2.5 MBytes) : Local copy, or directly from J. Phys. Chem. B, Copyright © American Chemical Society. See also the Supporting information file (3.3 Mbytes).

Regulation of nuclear receptors by their coactivators involves the recognition and binding of a specific sequence motif contained in the coactivator sequence. This motif is known as the nuclear receptor (NR) box and contains a conserved LxxLL subsequence, where L is leucine and x is any amino acid residue. Crystallographic studies have shown that the LxxLL motifs adopt an Î±-helical conformation when bound to their cognate nuclear receptors. Here we use an extensive set of folding molecular dynamics simulations to examine whether the Î±-helical conformation demonstrated by the LxxLL motifs in the bound state may represent a persistent structural preference of these peptides even in the absence of their cognate receptors. To this end, we have performed a grand total of 35 ÎŒs of adaptive tempering folding simulations of an NR-box-containing peptide derived from Drosophila's fushi tarazu segmentation gene product. Our simulationsâperformed using full electrostatics and an explicit representation of two different solvents (water and a TFE/water mixture)âclearly indicate the presence of a persistent helical preference of the LxxLL motif with a concomitant native-like structure and contacts between the motif's leucine residues. To lend further support to our findings, we compare the simulation-derived peptide dynamics with experimental NMR-derived nuclear Overhauser effect (NOE) measurements that had been previously obtained for the same peptide in the same two solvents. The comparison demonstrates a quantitative agreement between simulation and experiment with average upper bound NOE violations of less than 0.084 Ã, thus independently validating our main conclusion concerning the intrinsic preference of NR-box motifs to form helical structures even in the absence of their cognate receptors.

44. Fadouloglou, V.E., Balomenou, S., Aivaliotis, M., Kotsifaki, D., Arnaouteli, S., Tomatsidou, A., Efstathiou, G., Kountourakis, N., Miliara, S., Griniezaki, M., Tsalafouta, A., Pergantis, S.A., Boneca, I.G., Glykos, N.M., Bouriotis, V. & Kokkinidis*, M. (2017), "An unusual Î±-carbon hydroxylation of proline promotes active-site maturation", J. Am. Chem. Soc., 139, 5330â5337.
Electronic reprint (2.5 MBytes) : Local copy, or directly from J. Am. Chem. Soc., Copyright © American Chemical Society.

The full extent of proline (Pro) hydroxylation has yet to be established, as it is largely unexplored in bacteria. We describe here a so far unknown Pro hydroxylation activity which occurs in active sites of polysaccharide deacetylases (PDAs) from bacterial pathogens, modifying the protein backbone at the CÎ± atom of a Pro residue to produce 2-hydroxyproline (2-Hyp). This process modifies with high specificity a conserved Pro, shares with the deacetylation reaction the same active site and one catalytic residue and utilizes molecular oxygen as source for the hydroxyl group oxygen of 2-Hyp. By providing additional hydrogen bonding capacity, the Pro â 2-Hyp conversion alters the active site and enhances significantly deacetylase activity, probably by creating a more favorable environment for transition state stabilization. Our results classify this process as an active site "maturation", which is highly atypical by being a protein backbone modifying activity, rather than a side-chain modifying one.

43. Serafeim A.-P., Salamanos, G., Patapati, K.K. & Glykos*, N.M. (2016), "Sensitivity of Folding Molecular Dynamics Simulations to Even Minor Force Field Changes", J. Chem. Inf. Model., 56, 2035-2041.
Electronic reprint (2.6 MBytes) : Local copy, or directly from J. Chem. Inf. Model., Copyright © American Chemical Society. See also the Supporting information file.

We examine the sensitivity of folding molecular dynamics simulations on the choice between three variants of the same force field (the AMBER99SB force field and its ILDN, NMR-ILDN and STAR-ILDN variants). Using two different peptide systems (a marginally stable helical peptide and a Î²-hairpin) and a grand total of more than 20 ÎŒs of simulation time we show that even relatively minor force field changes can lead to appreciable differences in the peptide folding behavior.

42. Baltzis, A.S. & Glykos*, N.M. (2016), "Characterizing a partially ordered miniprotein through folding molecular dynamics simulations: Comparison with the experimental data", Prot. Sci., 25, 587â596.
Electronic reprint (3.3 MBytes) : Local copy, or directly from Protein Science, Copyright © Wiley Periodicals, Inc.

The villin headpiece helical subdomain (HP36) is one of the best known model systems for computational studies of fast-folding all-Î± miniproteins. HP21 is a peptide fragment âderived from HP36â comprising only the first and second helices of the full domain. Experimental studies showed that although HP21 is mostly unfolded in solution, it does maintain some persistent native-like structure as indicated by the analysis of NMR-derived chemical shifts. Here we compare the experimental data for HP21 with the results obtained from a 15 ÎŒs long folding molecular dynamics simulation performed in explicit water and with full electrostatics. We find that the simulation is in good agreement with the experiment and faithfully reproduces the major experimental findings, namely that (a) HP21 is disordered in solution with less that 10% of the trajectory corresponding to transiently stable structures, (b) the most highly populated conformer is a native-like structure with an RMSD from the corresponding portion of the HP36 crystal structure of less than 1Ã, (c) the simulation-derived chemical shifts âover the whole length of the trajectoryâ are in reasonable agreement with the experiment giving reduced Ï2 values of 1.6, 1.4 and 0.8 for the ÎÎŽ13CÎ±, ÎÎŽ13CO and ÎÎŽ13CÎ² secondary shifts respectively (becoming 0.8, 0.7, and 0.3 when only the major peptide conformer is considered), and finally, (d) the secondary structure propensity scores are in very good agreement with the experiment and clearly indicate the higher stability of the first helix. We conclude that folding molecular dynamics simulations can be a useful tool for the structural characterization of even marginally stable peptides.

41. Baltzis, A.S., Koukos, P.I. & Glykos*, N.M. (2015), "Clustering of molecular dynamics trajectories via peak-picking in multidimensional PCA-derived distributions", arXiv:1512.04024 [q-bio.BM].
Electronic reprint (826 KBytes) : Local copy, or directly from arXiv:1512.04024.

We describe a robust, fast, and memory-efficient procedure that can cluster millions of structures derived from molecular dynamics simulations. The essence of the method is based on a peak-picking algorithm applied to three- and five-dimensional distributions of the principal components derived from the trajectory and automatically supports both Cartesian and dihedral PCA-based clustering. The density threshold required for identifying isolated peaks (which correspond to discrete clusters) is determined through the application of a variance-explained criterion which allows for a completely automated clustering procedure with no user intervention. In this communication we describe the algorithm and present some of the results obtained from the application of the method as implemented in the molecular dynamics analysis programs carma, grcarma. and cluster5D. We conclude with a discussion of the limitations and possible pitfalls of this method.

40. Koukos, P.I. & Glykos*, N.M. (2014), "Folding Molecular Dynamics Simulations Accurately Predict the Effect of Mutations on the Stability and Structure of a Vammin-Derived Peptide", J. Phys. Chem. B, 118, 10076â10084.
Electronic reprint (3.7 MBytes) : Local copy, or directly from J. Phys. Chem. B, Copyright © American Chemical Society. See also the Supporting Information.

Folding molecular dynamics simulations amounting to a grand total of 4 ÎŒs of simulation time were performed on two peptides (with native and mutated sequences) derived from loop 3 of the vammin protein and the results compared with the experimentally known peptide stabilities and structures. The simulations faithfully and accurately reproduce the major experimental findings and show that (a) the native peptide is mostly disordered in solution, (b) the mutant peptide has a well-defined and stable structure, and (c) the structure of the mutant is an irregular Î²-hairpin with a non-glycine Î²-bulge, in excellent agreement with the peptideâs known NMR structure. Additionally, the simulations also predict the presence of a very small Î²-hairpin-like population for the native peptide but surprisingly indicate that this population is structurally more similar to the structure of the native peptide as observed in the vammin protein than to the NMR structure of the isolated mutant peptide. We conclude that, at least for the given system, force field, and simulation protocol, folding molecular dynamics simulations appear to be successful in reproducing the experimentally accessible physical reality to a satisfactory level of detail and accuracy.

39. Koukos, P.I. & Glykos*, N.M. (2014), "On the application of Good-Turing statistics to quantify convergence of biomolecular simulations", J. Chem. Inf. Model., 54, 209-217.
Electronic reprint (1.1 MBytes) : Local copy, or directly from J. Chem. Inf. Model., Copyright © American Chemical Society.

Quantifying convergence and sufficient sampling of macromolecular molecular dynamics simulations is more often than not a source of controversy (and of various ad hoc solutions) in the field. Clearly, the only reasonable, consistent and satisfying way to infer convergence (or otherwise) of a molecular dynamics trajectory must be based on probability theory. Ideally, the question we would wish to answer is the following : "What is the probability that a molecular configuration important for the analysis in hand has not yet been observed ?". Here we propose a method for answering a variant of this question by using the Good-Turing formalism for frequency estimation of unobserved species in a sample. Although several approaches may be followed in order to deal with the problem of discretizing the configurational space, for this work we use the classical RMSD matrix as a means to answering the following question : "What is the probability that a molecular configuration with an RMSD (from all other already observed configurations) higher than a given threshold has not actually been observed ?". We apply the proposed method to several different trajectories and show that the procedure appears to be both computationally stable and internally consistent. A free, open-source program implementing these ideas is immediately available for download via public repositories.

38. Koukos, P.I. & Glykos*, N.M. (2013), "grcarma: A Fully Automated Task-Oriented Interface for the Analysis of Molecular Dynamics Trajectories", J. Comput. Chem., 34, 2310-2312, Cover story for Vol.34, Issue 26.
Electronic reprint (929 KBytes) : Local copy, or directly from J. Comput. Chem., Copyright © Wiley Periodicals, Inc.

We report the availability of grcarma, a program encoding for a fully automated set of tasks aiming to simplify the analysis of molecular dynamics trajectories of biological macromolecules. It is a cross-platform, Perl/Tk-based front-end to the program carma and is designed to facilitate the needs of the novice as well as those of the expert user, while at the same time maintaining a user-friendly and intuitive design. Particular emphasis was given to the automation of several tedious tasks, such as extraction of clusters of structures based on dihedral and Cartesian principal component analysis, secondary structure analysis, calculation and display of root-meansquare deviation (RMSD) matrices, calculation of entropy, calculation and analysis of varianceâcovariance matrices, calculation of the fraction of native contacts, etc. The program is free-open source software available immediately for download.

37. Kontopoulos, D.-G. & Glykos*, N.M. (2013), "Pinda: A Web service for detection and analysis of intraspecies gene duplication events", Comput. Methods Programs Biomed., 111, 711-714.
Electronic reprint (729 KBytes) : Local copy, or directly from Comput. Methods Programs Biomed., Copyright © Elsevier Ltd.

We present Pinda, a Web service for the detection and analysis of possible duplications of a given protein or DNA sequence within a source species. Pinda fully automates the whole gene duplication detection procedure, from performing the initial similarity searches, to generating the multiple sequence alignments and the corresponding phylogenetic trees, to bootstrapping the trees and producing a Z-score-based list of duplication candidates for the input sequence. Pinda has been cross-validated using an extensive set of known and bibliographically characterized duplication events. The service facilitates the automatic and dependable identification of gene duplication events, using some of the most successful bioinformatics software to perform an extensive analysis protocol. Pinda will prove of use for the analysis of newly discovered genes and proteins, thus also assisting the study of recently sequenced genomes. The service's location is http://orion.mbg.duth.gr/Pinda. The source code is freely available via https://github.com/dgkontopoulos/Pinda/.

36. Georgoulia, P.S. & Glykos*, N.M. (2013), "On the Foldability of Tryptophan-Containing Tetra- and Pentapeptides: An Exhaustive Molecular Dynamics Study", J. Phys. Chem. B, 117, 5522â5532.
Electronic reprint (3.1 MBytes) : Local copy, or directly from J. Phys. Chem. B, Copyright © American Chemical Society. See also the Supporting Information.

Short peptides serve as minimal model systems to decipher the determinants of foldability due to their simplicity arising from their smaller size, their ability to echo protein-like structural characteristics, and their direct implication in force field validation. Here, we describe an effort to identify small peptides that can still form stable structures in aqueous solutions. We followed the in silico folding of a selected set of 8640 tryptophan-containing tetra- and pentapeptides through 15 210 molecular dynamics simulations amounting to a total of 272.46 ÎŒs using explicit representation of the solute and full treatment of the electrostatics. The evaluation and sorting of peptides is achieved through scoring functions, which include terms based on interatomic vector distances, atomic fluctuations, and rmsd matrices between successive frames of a trajectory. Highly scored peptides are studied further via successive simulation rounds of increasing simulation length and using different empirical force fields. Our method suggested only a handful of peptides with strong foldability prognosis. The discrepancies between the predictions of the various force fields for such short sequences are also extensively discussed. We conclude that the vast majority of such short peptides do not adopt significantly stable structures in water solutions, at least based on our computational predictions. The present work can be utilized in the rational design and engineering of bioactive peptides with desired molecular properties.

35. Patmanidis, I. & Glykos*, N.M. (2013), "As good as it gets? Folding molecular dynamics simulations of the LytA choline-binding peptide result to an exceptionally accurate model of the peptide structure", J. Mol. Graph. Model., 41, 68-71.
Electronic reprint (686 KBytes) : Local copy, or directly from J. Mol. Graph. Model., 41, 68-71, Copyright © Elsevier Inc.

Folding simulations of a choline-binding peptide derived from the Streptococcus pneumoniae LytA protein converged to a model of the peptide's folded state structure which is in outstanding agreement with the experimentally-determined structures, reaching values for the root mean squared deviation as low as 0.24 Ã for the peptide's backbone atoms and 0.65 Ã for all non-hydrogen atoms.

34. Fadouloglou, V.E., Kapanidou, M., Agiomirgianaki, A., Arnaouteli, S., Bouriotis, V., Glykos*, N.M. & Kokkinidis*, M. (2013), "Structure determination through homology modelling and torsion-angle simulated annealing: application to a polysaccharide deacetylase from Bacillus cereus", Acta Crystallogr., D69, 276-283.
Electronic reprint (2.4 MBytes) : Local copy, or directly from Acta Crystallogr., D69, 276-283, Copyright © International Union of Crystallography.

The structure of BC0361, a polysaccharide deacetylase from Bacillus cereus, has been determined using an unconventional molecular-replacement procedure. Tens of putative models of the C-terminal domain of the protein were constructed using a multitude of homology-modelling algorithms, and these were tested for the presence of signal in molecular-replacement calculations. Of these, only the model calculated by the SAM-T08 server gave a consistent and convincing solution, but the resulting model was too inaccurate to allow phase determination to proceed to completion. The application of slow-cooling torsion-angle simulated annealing (started from a very high temperature) drastically improved this initial model to the point of allowing phasing through cycles of model building and refinement to be initiated. The structure of the protein is presented with emphasis on the presence of a C(alpha)-modified proline at its active site, which was modelled as an alpha-hydroxy-L-proline.

33. Kokkinidis, M., Glykos, N.M. & Fadouloglou*, V.E. (2012), "Protein Flexibility and Enzymatic Catalysis", Adv. Protein Chem. Struct. Biol., 87, 181-218.
Electronic reprint (562 KBytes) : Local copy, or directly from Adv. Protein Chem. Struct. Biol., Copyright © Elsevier.

The dynamic nature of protein structures has been recognized, established, and accepted as an intrinsic fundamental property with major consequences to their function. Nowadays, proteins are considered as networks of continuous motions, which reflect local flexibility and a propensity for global structural plasticity. Proteinâprotein and proteinâsmall ligand interactions, signal transduction and assembly of macromolecular machines, allosteric regulation and thermal enzymatic adaptation are processes which require structural flexibility. In general, enzymes represent an attractive class among proteins in the study of protein flexibility and they can be used as model systems for understanding the implications of protein fluctuations to biological function. Flexibility of the active site is considered as a requirement for reduction of free energy barrier and acceleration of the enzymatic reaction while there is growing evidence which concerns the connection between flexibility and substrate turnover rate. Moreover, the role of conformational flexibility has been well established in connection with the accessibility of the active site, the binding of substrates and ligands, and release of products, stabilization and trapping of intermediates, orientation of the substrate into the binding cleft, adjustment of the reaction environment, etc.

32. Georgoulia, P.S. & Glykos*, N.M. (2011), "Using J-coupling constants for force field validation: Application to hepta-alanine", J. Phys. Chem. B, 115, 15221â15227.
Electronic reprint (3.6 MBytes) : Local copy, or directly from J. Phys. Chem. B, Copyright © American Chemical Society. See also the Supporting Information.

A computational solution to the protein folding problem is the holy grail of biomolecular simulation and of the corresponding force fields. The complexity of the systems used for folding simulations precludes a direct feedback between the simulations and the force fields, thus necessitating the study of simpler systems with sufficient experimental data to allow force field optimization and validation. Recent studies on short polyalanine peptides of increasing length (up to penta-alanine) indicated the presence of a systematic deviation between the experimental (NMR-derived) J-couplings and the great majority of biomolecular force fields, with the Ï² values for even the best-performing force fields being in the 1.4â1.8 range. Here we show that by increasing the number of residues to seven and by achieving convergence through an increase of the simulation time to 2 ÎŒs, we can identify one force field (the AMBER99SB force field, out of the three force fields studied) which when compared with the experimental J-coupling data (and for a specific set of Karplus-equation parameters and estimated J-coupling errors previously used in the literature) gave a value of Ï²=0.99, indicating that full statistical consistency between experiment and simulation is feasible. However, and as a detailed analysis of the effects of estimated errors shows, the Ï² values may be unsuitable as indicators of the goodness-of-fit of the various biomolecular force fields.

31. Patapati, K.K. & Glykos*, N.M. (2011), "Three Force Fields' Views of the 3₁₀ Helix", Biophys. J., 101, 1766-1771.
Electronic reprint (805 KBytes) : Local copy, or directly from Biophys. J., 101, 1766-1771. Copyright © 2011 Biophysical Society.

Slowly but steadily bibliographic evidence is accumulating that the apparent convergence of the various biomolecular force fields as evidenced from simulations of proteins in the folded state does not hold true for folding simulations. Here we add one more example to the growing list of peptides and proteins for which different force fields show irreconcilable differences in their folding predictions, even at such a fundamental level as that of a peptide's secondary structure. We show that for an undecamer peptide that is known from two independent NMR structure determinations to have a mainly 3₁₀-helical structure in solution, three mainstream biomolecular force fields give completely disparate predictions: The CHARMM force field (with the CMAP correction) predicts an outstandingly stable Î±-helical structure, in disagreement not only with the experimental structures, but also with experimental evidence obtained from circular dichroism. OPLS-AA shows an almost totally disordered peptide with the most frequently observed folded conformation corresponding to a Î²-hairpin-like structure, again in disagreement with all available experimental evidence. Only the AMBER99SB force field appears to qualitatively agree with not only the general structural characteristics of the peptide (on the account of both NMR- and CD-based experiments), but to also correctly predict some of the experimentally observed interactions at the level of side chains. Possible interpretations of these findings are discussed.

30. Glykos*, N.M. (2011), "On the application of structure-specific bulk-solvent models", Acta Crystallogr., D67, 739-741.
Electronic reprint (982 KBytes) : Local copy, or directly from Acta Cryst. D67, 739-741, Copyright © International Union of Crystallography.

It is often discussed, mainly in connection with the rather high macromolecular R factors, that the treatment of bulk solvent in macromolecular refinement may lack the detail needed for modelling the solvent environment of molecules as complex as proteins and nucleic acids. This line of thought directly leads to the hypothesis that improvements in the modelling of the bulk solvent may substantially improve the agreement between the experimental data and the crystallographic models. Here, part of this hypothesis is being tested through the construction, via molecular-dynamics simulations, of a highly detailed, physics-based, structure-specific and crystallographic data-agnostic model of the bulk solvent of a known crystal structure. The water-distribution map obtained from the simulation is converted (after imposing space-group symmetry) to a constant (but scalable) partial structure factor which is then added in a re-refinement of the crystal structure. Compared with the simple Babinet-based correction, a reduction of the totally cross-validated free R value by 0.3% is observed. The implications and possible interpretations of these results are discussed.

29. Glykos*, N.M. (2011), "The 11th Misconception?", A letter to the Editor, CBE Life Sci Educ, 10(1), 1-2.
Electronic reprint (123 KBytes) : Local copy, or directly from CBE Life Sci Educ, 10(1), 1-2, Copyright © Glykos under the Creative Commons Attribution License.

Biological sequences are an abstraction of an abstraction: In the first level, we substituted the complexity of a proper three-dimensional entity (like an amino acid residue) with a two dimensional chemical formula describing only composition and covalent bonding. In the second abstraction layer, we substituted these chemical formulas with single alphabet letters. And then we forgot about it, and started behaving as if sequences do exist, as if this artificial one-dimensionality is real. Sequence usage became so widespread, that not only we forgot that these one-dimensional strings of letters do not (and never did) exist, but we have started using them for dealing with problems (like protein folding) that by their nature defy this whole 'sequence' abstraction. Maybe, just maybe, we have had more than enough of 'sequences' ?

28. Patapati, K.K. & Glykos*, N.M. (2010), "Order through Disorder: Hyper-Mobile C-Terminal Residues Stabilize the Folded State of a Helical Peptide. A Molecular Dynamics Study", PLoS ONE, 5, e15290.
Electronic reprint (607 KBytes) : Local copy, or directly from PLos ONE, 5, e15290, Copyright © Patapati & Glykos under the Creative Commons Attribution License, See also the Supporting Information.

Conventional wisdom has it that the presence of disordered regions in the three-dimensional structures of polypeptides not only does not contribute significantly to the thermodynamic stability of their folded state, but, on the contrary, that the presence of disorder leads to a decrease of the corresponding proteins' stability. We have performed extensive 3.4 ÎŒs long folding simulations (in explicit solvent and with full electrostatics) of an undecamer peptide of experimentally known helical structure, both with and without its disordered (four residue long) C-terminal tail. Our simulations clearly indicate that the presence of the apparently disordered (in structural terms) C-terminal tail, increases the thermodynamic stability of the peptide's folded (helical) state. These results show that at least for the case of relatively short peptides, the interplay between thermodynamic stability and the apparent structural stability can be rather subtle, with even disordered regions contributing significantly to the stability of the folded state. Our results have clear implications for the understanding of peptide energetics and the design of foldable peptides.

27. Fadouloglou, V.E., Stavrakoudis, S., Bouriotis, V., Kokkinidis, M., & Glykos*, N.M. (2009), "Molecular Dynamics Simulations of BcZBP, A Deacetylase from Bacillus cereus: Active Site Loops Determine Substrate Accessibility and Specificity", J. Chem. Theory Comput., 5, 3299-3311.
Electronic reprint (7.3 MBytes) : Local copy, or directly from J. Chem. Theory Comput., 5, 3299-3311, Copyright © American Chemical Society.

BcZBP is an LmbE-like, homohexameric, zinc-dependent deacetylase from the opportunistic pathogen Bacillus cereus with three, thus far uncharacterized, homologues in B. anthracis. Although its specific substrate is still unknown, the enzyme has been shown to preferentially deacetylate N-acetylglucosamine and diacetylchitobiose via an active site based on a zinc-binding motif of the type HXDDXnH. In the crystal structure, the active site is located at a deep and partially blocked cleft formed at the interface between monomers related by the molecular 3-fold axis, although the major, in structural terms, building block of the enzyme is not the trimer, but the intertwined dimer. Here, we report results from a 50 ns molecular dynamics simulation of BcZBP in explicit solvent with full electrostatics and show that (i) the view of the intertwined dimer as the major structural and functional building block of this class of hexameric enzymes is possibly an oversimplification of the rather complex dynamics observed in the simulation, (ii) the most mobile (with respect to their atomic fluctuations) parts of the structure coincide with three surface loops surrounding the active site, and (iii) these mobile loops define the active site's accessibility, and may be implicated in the determination of the enzyme's specificity.

26. Fadouloglou, V.E., Bastaki, M.N., Ashcroft, A.E., Phillips, S.E.V., Panopoulos, N.J., Glykos*, N.M., & Kokkinidis*, M. (2009), "On the quaternary association of the type III secretion system HrcQB-C protein: Experimental evidence differentiates among the various oligomerization models", J. Struct. Biol., 166, 214-225.
Electronic reprint (3.7 MBytes) : Local copy, or directly from J. Struct. Biol., 166, 214-225, Copyright © Elsevier B.V.

The HrcQB protein from the plant pathogen Pseudomonas syringae is a core component of the bacterial type III secretion apparatus. The core consists of nine proteins widely conserved among animal and plant pathogens which also share sequence and structural similarities with proteins from the bacterial flagellum. Previous studies of the carboxy-terminal domain of HrcQB (HrcQB-C) and its flagellar homologue, FliN-C, have revealed extensive sequence and structural homologies, similar subcellular localization, and participation in analogous protein-protein interaction networks. It is not clear however whether the similarities between the two proteins extend to the level of quaternary association which is essential for the formation of higher-order structures within the TTSS. Even though the crystal structure of the FliN is a dimer, more detailed studies support a tetrameric donut-like association. However, both models, dimer and donut-like tetramer, are quite different from the crystallographic elongated dimer of dimers of the HrcQB-C. To resolve this discrepancy we performed a multidisciplinary investigation of the quaternary association of the HrcQB-C, including mass-spectrometry, electrophoresis in non-reductive conditions, gel filtration, glutaraldehyde cross-linking and small angle X-ray scattering. Our experiments indicate that stable tetramers of elongated shape are assembled in solution, in agreement with the results of crystallographic studies. Circular dichroism data are consistent with a dimer-dimer interface analogous to the one established in the crystal structure. Finally, molecular dynamics simulations reveal the relative orientation of the dimers forming the tetramers and the possible differences from that of the crystal structure.

25. Fadouloglou, V.E., Kokkinidis, M. & Glykos*, N.M. (2008), "Determination of protein oligomerization state: Two approaches based on glutaraldehyde crosslinking", Anal. Biochem., 373, 404-406.
Electronic reprint (605 KBytes) : Local copy, or directly from Anal. Biochem., 373, 404-406, Copyright © Elsevier B.V.

Many biochemical and biophysical methods can be used to characterize the oligomerization state of proteins. One of the most widely used is glutaraldehyde crosslinking, mainly because of the minimum equipment and reagents required. However, the crosslinking procedures currently in use are impaired by the low specificity of the reagent, which can chemically bond any two amino groups that are close in space. Thus, extensive and time-consuming investigation of the reaction conditions is usually required. Here we describe two approaches based on glutaraldehyde that readily give reliable results.

24. Mizas, Ch., Sirakoulis, G.Ch., Mardiris, V., Karafyllidis, I., Glykos, N.M. & Sandaltzopoulos*, R. (2008), "Reconstruction of DNA sequences using genetic algorithms and cellular automata: Towards mutation prediction ?", BioSystems, 92, 61-68.
Electronic reprint (574 KBytes) : Local copy, or directly from BioSystems, 92, 61-68, Copyright © Elsevier B.V.

Change of DNA sequence that fuels evolution is, to a certain extent, a deterministic process because mutagenesis does not occur in an absolutely random manner. So far, it has not been possible to decipher the rules that govern DNA sequence evolution due to the extreme complexity of the entire process. In our attempt to approach this issue we focus solely on the mechanisms of mutagenesis and deliberately disregard the role of natural selection. Hence, in this analysis, evolution refers to the accumulation of genetic alterations that originate from mutations and are transmitted through generations without being subjected to natural selection. We have developed a software tool that allows modelling of a DNA sequence as a one-dimensional cellular automaton (CA) with four states per cell which correspond to the four DNA bases, i.e. A, C, T and G. The four states are represented by numbers of the quaternary number system. Moreover, we have developed genetic algorithms (GAs) in order to determine the rules of CA evolution that simulate the DNA evolution process. Linear evolution rules were considered and square matrices were used to represent them. If DNA sequences of different evolution steps are available, our approach allows the determination of the underlying evolution rule(s). Conversely, once the evolution rules are deciphered, our tool may reconstruct the DNA sequence in any previous evolution step for which the exact sequence information was unknown. The developed tool may be used to test various parameters that could influence evolution. We describe a paradigm relying on the assumption that mutagenesis is governed by a near-neighbour-dependent mechanism. Based on the satisfactory performance of our system in the deliberately simplified example, we propose that our approach could offer a starting point for future attempts to understand the mechanisms that govern evolution. The developed software is open-source and has a user-friendly graphical input interface.

23. Fadouloglou, V.E., Deli, A., Glykos, N.M., Psylinakis, E., Bouriotis, V. & Kokkinidis*, M. (2007), "Crystal structure of the BcZBP, a zinc-binding protein from Bacillus cereus. Functional insights from structural data", FEBS J., 274, 3044-3054.
Electronic reprint (1.2 MBytes) : Local copy, or directly from FEBS J., 274, 3044-3054, Copyright © Blackwell Publishing, Inc.

Bacillus cereus is an opportunistic pathogenic bacterium closely related to B. anthracis, the causative agent of anthrax in mammals. A significant portion of the B. cereus chromosomal genes are common to B. anthracis, including genes which in B. anthracis code for putative virulence and surface proteins. B. cereus thus provides a convenient model organism for studying proteins potentially associated with the pathogenicity of the highly infectious B. anthracis. The zinc-binding protein of B. cereus, BcZBP, is encoded from the bc1534 gene which has three homologues to B. anthracis. The protein exhibits deacetylase activity with the N-acetyl moiety of the N-acetylglucosamine and the diacetylchitobiose and triacetylchitotriose. However, neither the specific substrate of the BcZBP nor the biochemical pathway have been conclusively identified. Here, we present the crystal structure of BcZBP at 1.8 A resolution. The N-terminal part of the 234 amino acid protein adopts a Rossmann fold whereas the C-terminal part consists of two beta strands and two alpha helices. In the crystal, the protein forms a compact hexamer, in agreement with solution data. A zinc binding site and a potential active site have been identified in each monomer. These sites have extensive similarities to those found in two known zinc-dependent hydrolases with deacetylase activity, MshB and LpxC, despite a low degree of amino acid sequence identity. The functional implications and a possible catalytic mechanism are discussed. (2ixd.pdb).

22. Glykos*, N.M. (2007), "On the application of molecular-dynamics simulations to validate thermal parameters and to optimize TLS-group selection for macromolecular refinement", Acta Crystallogr., D63, 705-713.
Electronic reprint (1.4 MBytes) : Local copy, or directly from Acta Cryst. D63, 705-713, Copyright © International Union of Crystallography.

Comparison of crystallographically determined and molecular dynamics simulation-derived parameters for a small (26 kDa) homotetrameric four-alpha-helical bundle protein revealed an unexpected pattern of similarities and differences between experiment and simulation. On one hand, the protein structure per se is exceptionally well preserved during the simulations, with a root-mean-square deviation between the Ca atoms of the crystal structure and the simulation-derived average structures of only 0.58 Angstrom, which is not very different from the expected coordinate error of the experimentally determined structure. On the other hand, comparison of the temperature factors showed a large discrepancy, with the experimental B-factors being approximately three times higher than the simulation-derived B-factors. Closer examination of this discrepancy appears to validate the molecular dynamics prediction and to implicate as its source static disorder at the crystalline state, as indicated by the strong diffuse scattering and pronounced anisotropy of the diffraction pattern of the protein crystals. A posteriori re-refinement of the structure using a new TLS parameterization scheme based on the results obtained from the simulations led to a further reduction of the R factor and the free R value by 0.4% and 0.8%, respectively, indicating that molecular-dynamics simulations have matured to the point that they can be used to aid the selection of TLS groups for macromolecular refinement.

21. Glykos*, N.M. (2006), "Carma: a molecular dynamics analysis program", J. Comput. Chem., 27, 1765-1768.
Electronic reprint (284 KBytes) : Local copy, or directly from Wiley InterScience, Copyright © Wiley InterScience.

A computer program has been developed to aid the analysis of molecular dynamics trajectories. The program is tuned for macromolecular large-scale problems and supports features such as removal of global translations-rotations of the solute, calculation of average distance maps and their corresponding standard deviations, calculation of the variance-covariance and cross-correlation matrices, and principal component analysis of trajectories with the added ability to create artificial trajectories based on selected eigenvectors. Limited graphics (trajectory viewing) capabilities are also available.

20. Glykos, N.M., Papanikolau, Y., Vlassi, M., Kotsifaki, D., Cesareni G. & Kokkinidis*, M. (2006), "Loopless Rop: Structure and Dynamics of an Engineered Homotetrameric Variant of the Repressor of Primer Protein", Biochemistry, 45, 10905-10919.
Electronic reprint (2.0 MBytes) : Local copy, or directly from the American Chemical Society, Copyright © American Chemical Society.

The repressor of primer (Rop) protein has become a steady source of surprises concerning the relationship between the sequences and the structures of several of its mutants and variants. Here we add another piece to the puzzle of Rop by showing that an engineered deletion mutant of the protein (corresponding to a deletion of residues 30-34 of the wild-type protein and designed to restore the heptad periodicity at the turn region) results in a complete reorganization of the bundle which is converted from a homodimer to a homotetramer. In contrast (and as previously shown), a two-residue insertion, which also restores the heptad periodicity, is essentially identical with wild-type Rop. The new deletion mutant structure is a canonical, left-handed, all-antiparallel bundle with a completely different hydrophobic core and distinct surface properties. The structure agrees and qualitatively explains the results from functional, thermodynamic, and kinetic studies which indicated that this deletion mutant is a biologically inactive hyperstable homotetramer. Additional insight into the stability and dynamics of the mutant structure has been obtained from extensive molecular dynamics simulations in explicit water and with full treatment of electrostatics. (1qx8.pdb).

19. Glykos*, N.M. (2005), "Qs v.1.3: a parallel version of Queen of Spades", J. Appl. Crystallogr., 38, 574-575.
Electronic reprint (65 KBytes) : Local copy, or directly from J. Appl. Crystallogr. 38, 574-575, Copyright © International Union of Crystallography.

The program Queen of Spades encodes an algorithm for a stochastic multidimensional approach to molecular replacement. The program has been shown to be capable of successfully locating solutions even in cases as complex as a 23-dimensional, 4-body problem. Recently, we extended our approach to tackle the full molecular replacement problem by allowing the possibility of using many different search models simultaneously, and showed that we could successfully locate solutions in the case of a 17-dimensional problem involving one DNA and two (different) protein search models. This multimodel, multidimensional approach does have its cost: with a few thousand unique reflections in a high symmetry space group and with more than two search models, a typical Qs run would take well over two to three weeks of CPU time on the fastest personal workstations. The way forward for such computationally intensive calculations is of course parallelisation. Here, I report the availability of a parallel version of Qs which is based on the Message Passing Interface (MPI) paradigm.

18. Glykos, N.M. & Kokkinidis*, M. (2004), "Structural Polymorphism of a Marginally Stable 4-alpha-Helical Bundle. Images of a Trapped Molten Globule ?", Proteins, 56, 420-425.
Electronic reprint (312 KBytes) : Local copy, or directly from Wiley InterScience, Copyright © Wiley InterScience.

The Repressor of primer (Rop) protein is the paradigm of a canonical (left-handed, all-anti-parallel) homodimeric 4-alpha-helical bundle. It is known (through the analysis of an orthorhombic crystal form) that a single alanine-to-proline amino-acid substitution at the turn region of Rop (the A31P mutant) markedly changes the topology of the protein which is converted to a right-handed, mixed parallel and antiparallel bundle. Here we report the structure of this mutant in a second (monoclinic) crystal form and show that although its topology remains unchanged, the differences between the two A31P crystal structures are unexpectedly large, with a root mean square deviation between equivalent Ca atoms of approximately 3A. Remarkably, a 3 ns molecular dynamics simulation of A31P which was initiated from the orthorhombic form crystal structure, sampled an ensemble of configurations rather similar to the structure seen in the monoclinic form. This finding suggests that the observed crystal structures may correspond to images of two conformers taken from a structurally heterogeneous population of molecules at equilibrium. Comparison of the A31P simulation with a 3 ns simulation of wild-type Rop indicated that the mutant is an inherently highly flexible molecule, both with respect to the relative placement of its helices and the malleability of its (loosely packed) hydrophobic core. Based on these findings, we propose that the A31P Rop mutant is an equilibrium molten globule and we attempt to interpret its thermodynamic properties based on this assumption. Furthermore, we present results from a 3 ns molecular dynamics simulation of a hypothetical structure which was constructed by artificially mutating the alanine at position 31 of the wild-type Rop structure to a proline. This hypothetical A31P structure appears to be significantly more stable than the experimentally determined one, leading us to propose that the observed A31P structure may correspond to a kinetically trapped molten globule. (1gmg.pdb).

17. Papanikolau, Y., Kotsifaki, D., Fadouloglou, V.E., Gazi, A.D., Glykos, N.M., Cesareni G. & Kokkinidis*, M. (2004), "Ionic strength reducers: an efficient approach to protein purification and crystallization. Application to two Rop variants.", Acta Crystallogr., D60, 1334-1337.
Electronic reprint (539 KBytes) : Local copy, or directly from Acta Cryst. D60, 1334-1337, Copyright © International Union of Crystallography.

Detailed knowledge of the influence of various parameters on macromolecular solubility is essential for crystallization. The concept of so-called ionic strength reducers provides insight into the changes in solubility induced by organic solvents and hydrophilic polymers in aqueous electrolytic solutions. A simple and efficient procedure is presented which exploits the properties of ionic strength reducers in the purification and crystallization of proteins. Using two designed variants of the Rop protein as model systems, superior crystals have been obtained compared with conventional techniques. This procedure is particularly useful in cases where excessive nucleation leads to the growth of a large number of tiny crystals that are useless for crystallographic analysis.

16. Glykos*, N.M. & Kokkinidis, M. (2004), "Molecular Replacement with multiple different models", J. Appl. Crystallogr., 37, 159-161.
Electronic reprint (1.1 MBytes) : Local copy, or directly from J. Appl. Crystallogr. 37, 159-161, Copyright © International Union of Crystallography.

Classical molecular replacement methods and the newer six-dimensional searches treat molecular replacement as a succession of sub-problems of reduced dimensionality. Due to their divide-and-conquer approach, these methods necessarily ignore (at least during their early stages) the very knowledge that a target crystal structure may comprise, for example, more than one copy of a search model, or, several models of different types. We have previously described an algorithm for a stochastic multidimensional molecular replacement search and showed that it can successfully locate solutions even in cases as complex as a 23-dimensional, 4-body search. The original description of the method only dealt with a special case of molecular replacement, namely with the problem of placing n copies of only one search model in the asymmetric unit of a target crystal structure. Here we present a natural generalisation of this algorithm to deal with the full molecular replacement problem, that is, with the problem of determining the orientations and positions of a total of n copies of m different models which are assumed to be present in the asymmetric unit of a target crystal structure. The generality of this approach is illustrated through its successful application to a 17-dimensional, 3-model problem involving one DNA and two protein molecules.

15. Fadouloglou, V.E., Tampakaki, A.P., Glykos, N.M., Bastaki, N., Hadden, J.M., Phillips, S.E., Panopoulos, N.J. & Kokkinidis*, M. (2004), "Structure of HrcQb-C, a conserved component of the bacterial type III secretion systems", Proc. Natl. Acad. Sci. USA, 101, 70-75.
Electronic reprint (629 KBytes) : Local copy, or directly from Proc. Natl. Acad. Sci. USA, 101, 70-75, Copyright © National Academy of Sciences USA.

Type III secretion systems enable plant and animal bacterial pathogens to deliver virulence proteins into the cytosol of eukaryotic host cells, causing a broad spectrum of diseases including bacteremia, septicemia, typhoid fever, and bubonic plague in mammals, and localized lesions, systemic wilting, and blights in plants. In addition, type III secretion systems are also required for biogenesis of the bacterial flagellum. The HrcQB protein, a component of the secretion apparatus of Pseudomonas syringae with homologues in all type III systems, has a variable N-terminal and a conserved C-terminal domain (HrcQB-C). Here, we report the crystal structure of HrcQB-C and show that this domain retains the ability of the full-length protein to interact with other type III components. A 3D analysis of sequence conservation patterns reveals two clusters of residues potentially involved in proteinprotein interactions. Based on the analogies between HrcQB and its flagellum homo- logues, we propose that HrcQB-C participates in the formation of a C-ring-like assembly. (1o9y.pdb).

14. Glykos*, N.M. & Kokkinidis, M. (2003), "Structure determination of a small protein through a 23-dimensional molecular replacement search", Acta Crystallogr., D59, 709-718.
Electronic reprint (3.8 MBytes) : Local copy, or directly from Acta Cryst. D59, 709-718, Copyright © International Union of Crystallography.

The crystal structure of a 4-alpha-helical bundle protein has been determined through the application of a 23-dimensional molecular replacement search performed with a stochastic method. The search model for the calculation was a 26 residue-long poly-alanine helix amounting to less than 13% of the total number of atoms in the asymmetric unit of the target crystal structure. The crystal structure determination procedure is presented in detail, with emphasis on the molecular replacement calculations (1gmg.pdb).

13. Dennis, C.A., Glykos, N.M., Parsons, M.R. & Phillips*, S.E.V. (2002), "The structure of AhrC, the arginine repressor/activator protein from Bacillus subtilis", Acta Crystallogr., D58, 421-430.
Electronic reprint (1.6 MBytes) : Local copy, or directly from Acta Cryst. D58, 421-430, Copyright © International Union of Crystallography.

In the gram positive bacterium Bacillus subtilis the concentration of the amino acid L-arginine is controlled by the transcriptional regulator AhrC. The hexameric AhrC protein binds in an L-arginine-dependent manner to pseudo-palindromic operators within the prometer regions of arginine biosynthesic and catabolic gene clusters. AhrC binding results in the repression of transcription of biosynthetic genes and in the activation of transcription of catabolic genes. We have determined the crystal structure of AhrC at 2.7A resulution. Each sununit of the protein has two domains. The C-terminal domains are arranged with 32 point group symmetry and mediate the major inter-subunit interactions. The N-terminal domains are located around this core, where they lie in weakly associated pairs but do not obey strict symmetry. A structural comparison of AhrC with the arginine repressor from the thermophile Bacillus stearothermophilus reveals close similarity in regions implicated in L-arginine binding and DNA recognition but also some striking sequence differences, especially within the C-terminal oligomerisation domain, which may contribute to the different thermostabilities of the proteins. Comparison of the crystal structure of AhrC with a 30A resolution model obtained by combining X-ray structure factor amplitudes with phases derived from electron microscopic analyses of AhrC crystals confirms the essential accuracy of the earlier model and suggests that such an approach may be more widely useful for obtaining low resolution phase information (1f9n.pdb).

12. Glykos*, N.M. & Kokkinidis, M. (2001), "Multidimensional Molecular Replacement", Acta Crystallogr., D57, 1462-1473.
Electronic reprint (1.9 MBytes) : Local copy, or directly from Acta Cryst. D57, 1462-1473, Copyright © International Union of Crystallography.
The ECM 20 (Krakow, 2001) presentation is also available for download (Powerpoint presentation, 1.6 MB).

A method is described which attempts to simultaneously and independently determine the positional and orientational parameters of all molecules present in the asymmetric unit of a target crystal structure. This is achieved through a reverse Monte Carlo optimisation of a suitable statistic (like the R-factor or the linear correlation coefficient between the observed and calculated amplitudes of the structure factors) in the 6n-dimensional space defined by the rotational and translational parameters of the n search models. Results from the application of this stochastic method ---obtained with a space group general computer program which has been developed for this purpose--- indicate that with present-day computing capabilities the method may successfully be applied to molecular replacement problems for which the target crystal structure contains up to three molecules per asymmetric unit. It is also shown that the method may be useful in cases where the assumption of topological segregation of the self and cross vectors in the Patterson function is violated (as may happen, for example, in closely packed crystal structures).

11. Fadouloglou, V.E., Glykos, N.M. & Kokkinidis*, M. (2001), "Side-chain conformations in 4-alpha-helical bundles", Protein Engng., 14, 321-328.
Electronic reprint (385 KBytes) : Local copy, or directly from Protein Engng. 14, 321-328, Copyright © Oxford University Press.

The distribution of the chi1, chi2 dihedral angles in a data set consisting of twelve unrelated 4-alpha-helical bundle proteins has been determined and compared with that observed in globular proteins. The analysis suggests that for this tertiary motif : (i) the side-chain conformations are limited to only a subset of the conformations observed in globular proteins and are more constrained that side chains in helical regions of globular proteins. (ii) The side chains of Aspartic acid and Asparagine adopt occasionally a new, topology-specific rotamer. (iii) The rotamer preferences of Tyrosine and Isoleucine depend on whether they are located in the hydrophobic core of the bundle or in a more exposed position. (iv) Naturally occuring mutations in the hydrophobic core of 4-alpha-helical bundles follow a pattern that is consistent with the notion that the mutated and wild-type residues have at least one of their most highly populated rotamers in common. These findings indicate a relationship between protein topology and preferred side-chains conformations.

10. Fadouloglou, V.E., Glykos*, N.M. & Kokkinidis, M. (2000), "A fast and inexpensive procedure for drying polyacrylamide gels.", Anal. Biochem., 287, 185-186.
Electronic reprint (69 KBytes) : Local copy, or directly from Anal. Biochem. 287, 185-186, Copyright © Academic Press.

A simple procedure for drying polyacrylamide gels is described. This only involves soaking the gel twice in low grade ethanol : The gel is placed in a petri dish containing 5-10 gel volumes of low grade ethanol and stirred for 10-15 minutes. After that time, the ethanol solution is refreshed and soaking continues for 5 minutes. During this second soak the gel becomes opaque, dehydrates and shrinks uniformly (without cracking) by a factor of about 35-40%. In the final step, the gel is removed from the ethanol solution, placed on a hard (non-adhesive) surface, ethanol is allowed to evaporate from its top surface, and then it is covered with a glass plate to avoid curling during the final stages of ethanol evaporation. After few hours the gel is ready for storage.

9. Spyridaki, A., Glykos, N.M., Kotsifaki, D., Fadouloglou, V. & Kokkinidis*, M. (2000), "Crystallization and diffraction to ultrahigh resolution (0.8A) of a designed variant of the Rop protein.", Acta Crystallogr., D56, 1015-1016.
Electronic reprint (237 KBytes) : Local copy, or directly from Acta Cryst. D56, 1015-1016, Copyright © International Union of Crystallography.

$Diffraction pattern of an RM7 crystal, courtesy of Spyridaki Aspasia, 60 Kbytes$

The Rop protein is the paradigm of a highly regular 4-a-helix bundle, and as such it has been subject to numerous structural and mutagenesis studies. Crystals of a designed Rop variant which establishes a continuous heptad pattern through the bend region have been obtained by a combination of vapour diffusion and seeding techniques. The crystals diffract to ultrahigh (0.8A) resolution using synchrotron radiation and cryogenic conditions.

8. Glykos, N.M. & Kokkinidis*, M. (2000), "On the distribution of the bulk solvent correction parameters", Acta Crystallogr., D56, 1070-1072.
Electronic reprint (384 KBytes) : Local copy, or directly from Acta Cryst. D56, 1070-1072, Copyright © International Union of Crystallography.

The distribution of the bulk solvent correction parameters (Bsol, ksol) -as determined with an exponential scaling algorithm based on Babinet's principle- for 219 crystal structures deposited with the Protein Data Bank is presented. The observed distribution strongly suggests that the two parameters are not independent, and a reasonable agreement with the experimental data could be obtained through the application of a simple exponential function. Possible interpretations of this finding are discussed.

7. Glykos*, N.M. & Kokkinidis, M. (2000), "GraphEnt : a maximum entropy program with graphics capabilities.", J. Appl. Crystallogr., 33, 982-985.
Electronic reprint (584 KBytes) : Local copy, or directly from J. Appl. Cryst. 33, 982-985, Copyright © International Union of Crystallography.

A maximum entropy formalism aiming at the production of a "maximally non-committal" map is a standard method in fields of science like radioastronomy, but a rare exception in both X-ray crystallography and electron microscopy (or crystallography). This is rather unfortunate, given the wealth of information that a MAXENT map can reveal, especially when the map itself is the end product (for example, low resolution electron or potential density maps, Patterson functions, deformation maps). The program GraphEnt attempts to automate the procedure of calculating maximum entropy maps, with emphasis on the calculation of difference Patterson functions for macromolecular crystallographic problems, while providing a useful graphical output of the current stage of the calculation.

6. Andreeva, A.E., Borissova, B.E., Mironova, R., Glykos, N.M., Kotsifaki, D., Ivanov, I., Krysteva, M. & Kokkinidis*, M. (2000), "Crystallization of type I chloramphenicol acetyltransferase : An approach based on the concept of ionic strength reducers", Acta Crystallogr., D56, 101-103.
Electronic reprint (319 KBytes) : Local copy, or directly from Acta Cryst. D56, 101-103, Copyright © International Union of Crystallography.

$Diffraction pattern of a CAT crystal (plus inset), 91 Kbytes$

Chloramphenicol acetyltransferase (CAT) is responsible for bacterial resistance to chloramphenicol. It catalyzes inactivation of the antibiotic by acetyl group transfer from acetyl CoA to one or both hydroxyl groups of chloramphenicol. Type I CAT possesses some unique properties which are not observed in other CAT variants. Type I CAT overexpressed in E.coli was purified and crystals with a resolution limit of 2.22 Å have been obtained using a novel procedure which is based on the concept of the "ionic strength reducers". The crystals have the symmetry of spacegroup P1 and the unit cell dimensions are a=96.5, b=113.9, c=114.2, alpha=119.9, beta=94.1 and gamma=98.6 degrees. These dimensions are consistent with four to six trimers per unit cell corresponding to a solvent fraction ranging from 65 to 47%.

5. Glykos*, N.M. & Kokkinidis, M. (2000), "A stochastic approach to Molecular Replacement", Acta Crystallogr., D56, 169-174.
Electronic reprint (415 KBytes) : Local copy, or directly from Acta Cryst. D56, 169-174, Copyright © International Union of Crystallography.
The CCP4 2001 Study Weekend presentation is also available for download (Powerpoint presentation, 1 MB).
The ECM 20 (Krakow, 2001) presentation is also available for download (Powerpoint presentation, 1.6 MB).

The classical approach to the problem of placing n copies of a search model in the asymmetric unit of a target crystal structure, is to divide this 6n-dimensional optimisation problem into a succession of 3-dimensional searches (rotation function followed by translation function searches for each of the models). Here it is shown that a structure determination method based on a reverse Monte Carlo minimisation of the conventional crystallographic R-factor in the 6n-dimensional space defined by the rotational and translational parameters of the n molecules, is both feasible and practical, at least for small n. Because all parameters of all molecules are determined simultaneously, this algorithm should improve the signal-to-noise ratio in difficult cases involving high crystallographic/non-crystallographic symmetry in tightly packed crystal forms. Preliminary results from the application of this method -obtained with a space group general computer program which has been developed for this purpose- are presented.

4. Glykos, N.M. & Kokkinidis*, M. (1999), "Meaningful refinement of poly-alanine models using rigid-body simulated annealing : application to the structure determination of the A31P Rop mutant.", Acta Crystallogr., D55, 1301-1308.
Electronic reprint (1.7 MBytes) : Local copy, or directly from Acta Cryst. D55, 1301-1308, Copyright © International Union of Crystallography.

Conventional least-squares or simulated annealing refinement of even correctly positioned poly-alanine models of a target structure, results to a systematic distortion of the molecular geometry and to a concomitant increase of the mean phase difference from the correct phase set. Here it is shown that iterative rigid-body simulated annealing refinement of poly-alanine models employing successively fewer residues per rigid body (down to one alanine residue per body) at a very high initial temperature (of the order of To=10000 K) and with the geometric energy terms switched on, not only preserves the geometry of the model, but can also converge to an essentially correct poly-alanine trace of the target structure even when the starting model deviates systematically and significantly from the sought structure. As an example of the application of the method we present details of the structure determination of the alanine-31 to proline mutant of the Rop protein, where an initial, roughly positioned poly-alanine model (giving an average phase difference of 78.2 degrees from the final phase set) was successfully refined against a 1.8A resolution native data set, leading to an essentially correct model of the main chain with an average displacement of its atomic positions from the final model of 0.275A. The phases calculated from this refined poly-alanine model had an average difference of 43.8 degrees from the final phase set (corresponding to a mean figure of merit of 0.72) and gave a readily interpretable electron density map.

3. Glykos*, N.M. (1999), "Pepinsky's Machine : an interactive, graphics-based Fourier synthesis program with applications in teaching and research.", J. Appl. Crystallogr., 32, 821-823.
Electronic reprint (660 KBytes) : Local copy, or directly from J. Appl. Cryst. 32, 821-823, FREE ITEM, Copyright © International Union of Crystallography.

A computer program has been developed which given a set of structure factor amplitudes for any centrosymmetric plane group, displays the amplitude-weighted reciprocal lattice plane and allows the user to interactively assign and modify the phases of the structure factors, while observing the effect of these changes on the corresponding electron density function. The program has the added feature of being able to calculate and interactively display the electron density maps corresponding to all phase combinations of a user-defined subset of structure factors. The application of the program in both crystallographic teaching and research are discussed.

2. Glykos, N.M., Cesareni, G. & Kokkinidis*, M. (1999), "Protein plasticity to the extreme : Changing the topology of a 4-alpha-helical bundle with a single amino-acid substitution.", Structure 7, 597-603.
Electronic reprint (2.2 MBytes) : Local copy, or directly from ScienceDirect, Copyright © Current Biology Publications.

The general experience from structural studies of single amino-acid substituted mutant proteins, is that the effect of mutation is rather localised and minor. Here we provide a counter-example to this statement by showing that a single alanine to proline substitution in the turn region of a 4-alpha-helical protein leads to a complete reorganisation of the whole molecule which is converted from the canonical left-handed all-antiparallel form, to a right-handed, mixed parallel and antiparallel bundle, which to the best of our knowledge and belief represents a novel topological motif for this class of proteins. Our results suggest a possible new mechanism for the creation and evolution of protein folds, show the importance of the loop regions in determining the allowable folding pathways and illustrate the malleability of protein structure (1b6q.pdb).

1. Glykos, N.M., Holzenburg, A.K.H. & Phillips*, S.E.V. (1998), "Low resolution structural characterisation of the Arginine repressor/activator from Bacillus subtilis : A combined X-ray crystallographic and electron microscopical approach", Acta Crystallogr., D54, 215-225, and Acta Crystallogr., D54, 707-707.
Electronic reprint (2.6 MBytes) : Local copy, or directly from Acta Cryst. D54, 215-225, Copyright © International Union of Crystallography.

Attempts to determine the X-ray crystal structure of the intact homohexameric arginine repressor/activator from B. subtilis have so far been unsuccessful. The major problem appears to be the lack of an isomorphous heavy atom derivative with a manageable number of substitution sites. Here it is shown how electron microscopy of thin three-dimensional crystals, the same as those used for the X-ray crystallographic studies, made it possible (i) to obtain experimental support for some conclusions drawn on the basis of X-ray data alone, (ii) to determine the low resolution distribution of electron density in several different crystallographic projections, and, (iii) to obtain a tentative low resolution model of the whole hexamer.

0. Glykos, N.M. (1995), "Structural studies of the arginine repressor/activator from Bacillus subtilis", PhD thesis, Thesis advisor Prof Simon E.V. Phillips, Astbury Department of Biophysics, University of Leeds.
Electronic reprint (7.7 MBytes)

In the presence of L-Arginine, AhrC --the Arginine-dependent Repressor/Activator from Bacillus subtilis-- represses the transcription of the genes encoding the anabolic and activates those encoding the catabolic enzymes of arginine metabolism. AhrC is a homohexamer of total molecular mass 105 kDa. It shows no homology to any of the characterised DNA-binding motifs or DNA-binding proteins with the exception of ArgR, the Arginine Repressor from Escherichia coli. ArgR does not act as a transcription activator but it has been shown to be a necessary accessory protein for the resolution --through site-specific recombination-- of multimers of the ColE1 plasmid. Although the two proteins share only 29% identity and are from such taxonomically distinct prokaryotes, AhrC can complement E. coli ArgR- strains both in the regulation of Arginine metabolism and the resolution of the ColE1 plasmid. This thesis describes our attempts to determine the crystal structure of AhrC. Three different crystal forms have been produced and characterised. Useful derivatives have been prepared for two of these forms but the determination of their heavy atom structures proved impossible. An attempt to determine the low resolution structure of AhrC using Electron Microscopy has been unsuccessful. Molecular Replacement using as a search model the crystal structure of the hexameric core fragment of ArgR also failed to give a convincing solution.

Structural and computational biologyNicholas M. Glykos' group